FYI...
http://www.smartofficenews.com.au/Computing/Platforms_And_Applications?artic... e=/Computing/Platforms%20And%20Applications/News/E5T7U6H8&page=1
Is this a threat to Linux?
On Tue, 2005-09-27 at 06:28 +0200, Fajar Priyanto wrote:
FYI...
http://www.smartofficenews.com.au/Computing/Platforms_And_Applications?artic... e=/Computing/Platforms%20And%20Applications/News/E5T7U6H8&page=1
Is this a threat to Linux?
No.
On Mon, 2005-09-26 at 21:37 -0700, Michael A. Peters wrote:
On Tue, 2005-09-27 at 06:28 +0200, Fajar Priyanto wrote:
FYI...
http://www.smartofficenews.com.au/Computing/Platforms_And_Applications?artic... e=/Computing/Platforms%20And%20Applications/News/E5T7U6H8&page=1
Is this a threat to Linux?
No.
Ya know...I had this big wordy response typed out in my Evolution composition window, but you said everything I wanted to say in one simple word and statement. :)
I probably just need more caffeine...
On Tue, 2005-09-27 at 00:41, Peter Gordon wrote:
On Mon, 2005-09-26 at 21:37 -0700, Michael A. Peters wrote:
On Tue, 2005-09-27 at 06:28 +0200, Fajar Priyanto wrote:
FYI...
http://www.smartofficenews.com.au/Computing/Platforms_And_Applications?artic... e=/Computing/Platforms%20And%20Applications/News/E5T7U6H8&page=1
Is this a threat to Linux?
No.
Ya know...I had this big wordy response typed out in my Evolution composition window, but you said everything I wanted to say in one simple word and statement. :)
I probably just need more caffeine...
Actually the important part to note from the article is that they have reorganized into three business units. Watch for Microsoft to spin off one or more of those units in coming years.......
On Tue, 2005-09-27 at 09:03 -0400, Scot L. Harris wrote:
Actually the important part to note from the article is that they have reorganized into three business units. Watch for Microsoft to spin off one or more of those units in coming years.......
Did You hear about Linus's selling of the system to MS? Does anybody know why except for money he did it? By the way, what is the best Linux/Open Source Community news site where subscribing is possible?
Thomas Cameron wrote:
On Tue, 2005-09-27 at 22:59 +0700, Strong wrote:
Did You hear about Linus's selling of the system to MS?
What are you talking about?
Thomas
I had to search for google on this one. Fortunately, nothing regarding a sellout of any kind was found. I did find an interesting article regarding Linux, Mosaic (freeware what then became Netscape), the Wordperfect 6 for Linux which was capable of running Windows 3.1 apps. I believe the article is from 1997 timeframe though.
http://www.metroactive.com/papers/metro/05.08.97/cover/linus-9719.html
Jim
On Sun, 2005-10-02 at 10:20 -0400, Jim Cornette wrote:
Wordperfect 6 for Linux which was capable of running Windows 3.1 apps. I believe the article is from 1997 timeframe though.
Is wordperfect not a text processor? How does it run an application?!
Strong wrote:
On Sun, 2005-10-02 at 10:20 -0400, Jim Cornette wrote:
Wordperfect 6 for Linux which was capable of running Windows 3.1 apps. I believe the article is from 1997 timeframe though.
Is wordperfect not a text processor? How does it run an application?!
It's a word processor; as I recall one could use its Windows printer drivers on Linux: I conclude they're not drives as system hackers would understand the term.
On Sat, 2005-11-26 at 01:22 +0600, Strong wrote:
On Sun, 2005-10-02 at 10:20 -0400, Jim Cornette wrote:
Wordperfect 6 for Linux which was capable of running Windows 3.1 apps. I believe the article is from 1997 timeframe though.
Is wordperfect not a text processor? How does it run an application?!
The same way Microsoft Word does it. It makes system calls. Often by using macros.
-- Best regards, Strong.
Strong wrote:
On Sun, 2005-10-02 at 10:20 -0400, Jim Cornette wrote:
Wordperfect 6 for Linux which was capable of running Windows 3.1 apps. I believe the article is from 1997 timeframe though.
Is wordperfect not a text processor? How does it run an application?!
The concept was to load wordperfect under Linux. Also WordPerfect was working on a program that was able to run Microsoft applications under Linux. MicroSoft bought WordPerfect out and closed development on the project and stripped out the code WordPerfect developed for the feat. I was going to buy it when it came out. I believe there was a version of WordPerfect Linux that actually made it to the stores. I do not know whatever happened to the information on the web. I'll try WordPerfect Linux to see what comes up. What is a competent search engine? Google seems to be falling apart. It could not find the sun in midday now.
Jim
On Fri, Nov 25, 2005 at 10:52:27PM -0500, Jim Cornette wrote:
The concept was to load wordperfect under Linux. Also WordPerfect was working on a program that was able to run Microsoft applications under Linux. MicroSoft bought WordPerfect out and closed development on the project and stripped out the code WordPerfect developed for the feat.
This isn't quite accurate. Corel made use of the open-source Wine project to port their code to Linux (using it in its "Wine-is-not-an-emulator" role). They did make contributions back to the open source project, but I'm not sure how extensive they were.
I was going to buy it when it came out. I believe there was a version of WordPerfect Linux that actually made it to the stores. I do not know
It was a free download; I'm not sure if it was actually also available in physical stores for money.
whatever happened to the information on the web. I'll try WordPerfect Linux to see what comes up.
Yes, good idea.
What is a competent search engine? Google seems to be falling apart. It could not find the sun in midday now.
Well, try Google before complaining. :)
Searching for "wordperfect linux" (without the quotes) on Google brings up a very informative FAQ, including links to download sites.
Matthew Miller wrote:
On Fri, Nov 25, 2005 at 10:52:27PM -0500, Jim Cornette wrote:
The concept was to load wordperfect under Linux. Also WordPerfect was working on a program that was able to run Microsoft applications under Linux. MicroSoft bought WordPerfect out and closed development on the project and stripped out the code WordPerfect developed for the feat.
This isn't quite accurate. Corel made use of the open-source Wine project to port their code to Linux (using it in its "Wine-is-not-an-emulator" role). They did make contributions back to the open source project, but I'm not sure how extensive they were.
I was going to buy it when it came out. I believe there was a version of WordPerfect Linux that actually made it to the stores. I do not know
It was a free download; I'm not sure if it was actually also available in physical stores for money.
whatever happened to the information on the web. I'll try WordPerfect Linux to see what comes up.
Yes, good idea.
What is a competent search engine? Google seems to be falling apart. It could not find the sun in midday now.
Well, try Google before complaining. :)
Searching for "wordperfect linux" (without the quotes) on Google brings up a very informative FAQ, including links to download sites.
I tried using Google and it did not net anything close to what WordPerfect product development did. This excerpt was found in news through Google.
*News* 1-Jan-2002 All prior Corel Linux links now point to Xandros http://www.xandros.net/ 2-Oct-2000 Microsoft invests $135 Million and establishes a strategic alliance http://www.theosfiles.com/os_linux/ospg_Linux_corel.htm
Myself, I am getting a little less confident on Google searches that I have tried on both historic and recent event searches. It is not as good as it once seemed to be.
As far as the suppression of technologies by large corporations, it goes on in efficient transportation as well as software technologies.
WordPerfect Linux or Corel Linux are outlawed or eliminated. I guess focus on what heppens with current technologies and what government regulations (bought by corporations) are put on Multimedia, OS and other items is the better item to keep in focus.
Jim
On Fri, 25 Nov 2005, Jim Cornette wrote:
Strong wrote:
On Sun, 2005-10-02 at 10:20 -0400, Jim Cornette wrote:
Wordperfect 6 for Linux which was capable of running Windows 3.1 apps. I believe the article is from 1997 timeframe though.
Is wordperfect not a text processor? How does it run an application?!
The concept was to load wordperfect under Linux. Also WordPerfect was working on a program that was able to run Microsoft applications under Linux. MicroSoft bought WordPerfect out and closed development on the project and stripped out the code WordPerfect developed for the feat. I was going to buy it when it came out. I believe there was a version of WordPerfect Linux that actually made it to the stores. I do not know whatever happened to the information on the web. I'll try WordPerfect Linux to see what comes up. What is a competent search engine? Google seems to be falling apart. It could not find the sun in midday now.
Word Perfect, when it was a standalone company, had a version of Word Perfect (the software) for various Unix OS's, including Linux if I recall correctly. Then Novell bought the company, and in the shakeup the product fell behind MS Word/Office, although it became an Office suit in its own right about that time. Under the original Caldera Linux product (rebranded RH 3.25 or some such thing) I ran the Linux version of the office suite. Corel bought the product, bringing more damage to the table as things got moved yet again. As I recall, Corel was (and started?? did??) compiling the whole office suite to run with Wine binaries, so in theory you could either run under Windows or under Wine with the same binaries. There was also some effort, as I recall, to port portions/the whole mess to Java, but I think that was pretty still born.
FWIW, IMHO, Novell killed a great product and the last real agressive customer sevice organization in consumer computing. The times I called WP help, I may have waited online for some time, but the techs I spoke with _knew_ the product, followed up, and spoke a brand of english I could understand. I was very sorry to see that service go away.
Jim
tfreeman@intel.digichem.net wrote:
On Fri, 25 Nov 2005, Jim Cornette wrote:
Strong wrote:
On Sun, 2005-10-02 at 10:20 -0400, Jim Cornette wrote:
Wordperfect 6 for Linux which was capable of running Windows 3.1 apps. I believe the article is from 1997 timeframe though.
Is wordperfect not a text processor? How does it run an application?!
The concept was to load wordperfect under Linux. Also WordPerfect was working on a program that was able to run Microsoft applications under Linux. MicroSoft bought WordPerfect out and closed development on the project and stripped out the code WordPerfect developed for the feat. I was going to buy it when it came out. I believe there was a version of WordPerfect Linux that actually made it to the stores. I do not know whatever happened to the information on the web. I'll try WordPerfect Linux to see what comes up. What is a competent search engine? Google seems to be falling apart. It could not find the sun in midday now.
Word Perfect, when it was a standalone company, had a version of Word Perfect (the software) for various Unix OS's, including Linux if I recall correctly. Then Novell bought the company, and in the shakeup the product fell behind MS Word/Office, although it became an Office suit in its own right about that time. Under the original Caldera Linux product (rebranded RH 3.25 or some such thing) I ran the Linux version of the office suite. Corel bought the product, bringing more damage to the table as things got moved yet again. As I recall, Corel was (and started?? did??) compiling the whole office suite to run with Wine binaries, so in theory you could either run under Windows or under Wine with the same binaries. There was also some effort, as I recall, to port portions/the whole mess to Java, but I think that was pretty still born.
FWIW, IMHO, Novell killed a great product and the last real agressive customer sevice organization in consumer computing. The times I called WP help, I may have waited online for some time, but the techs I spoke with _knew_ the product, followed up, and spoke a brand of english I could understand. I was very sorry to see that service go away.
Thanks for the info. Searches for info revealed timetables where the UNIX, VAX, Amiga, and other hardware types were mentioned. There were mentions of their JAVA attempts also in that timeframe. Your recollection helped me piece together some of the events.
I'll lookup Caldera Linux, which was found on some of the searches. (with many different search engines)
Jim
On Sat, Nov 26, 2005 at 01:27:58AM -0500, Jim Cornette wrote:
tfreeman@intel.digichem.net wrote:
On Fri, 25 Nov 2005, Jim Cornette wrote:
Strong wrote: The concept was to load wordperfect under Linux. Also WordPerfect was working on a program that was able to run Microsoft applications under Linux. MicroSoft bought WordPerfect out and closed development on the project and stripped out the code WordPerfect developed for the feat. I was going to buy it when it came out. I believe there was a version of WordPerfect Linux that actually made it to the stores. I do not know whatever happened to the information on the web. I'll try WordPerfect Linux to see what comes up. What is a competent search engine? Google seems to be falling apart. It could not find the sun in midday now.
Microsoft took a pretty decent stake in Corel. I think it was much along the lines of their investment in Apple years ago. They needed a competitor. WordPerfect is still in active development, but I think on Windows only.
I personally purchased WordPerfect 8 for Linux retail. IIRC, it went for $50.
It was also available as a free download, sans the font installer. It dated from the late 90s. Was a LibC 5 app. The last distro I could run it under was RH 6.2 as it still had the LibC 5 compatability RPM. You could even get the download version in a Word Perfect for Linux Bible book as well. There was a very active community who used it as well.
They did market the Wine compiled version for a while. But the write-once and run on Windows and Linux effort failed, even with Wine. Though they did give back quite a bit of work to the Wine project.
At the time it was, IMHO, the best word processor on Linux. Version 8 was a native Linux app ported by a company for Corel. The only weird this was that it had it's own printing engine. But you could have it call a different application, like Kprint to do other stuff. It also only used PostScript fonts.
Word Perfect, when it was a standalone company, had a version of Word Perfect (the software) for various Unix OS's, including Linux if I recall correctly. Then Novell bought the company, and in the shakeup the product fell behind MS Word/Office, although it became an Office suit in its own right about that time. Under the original Caldera Linux product (rebranded RH 3.25 or some such thing) I ran the Linux version of the office suite.
The office suite from Novell was PerfectOffice and was around the WP 6.2 days. I worked with it around 96-98 or so.
Caldera's distro was OpenLinux and was derived from RH back in the day. (It was actually my first distro I ever tried (and failed on.)
Corel bought the product, bringing more damage to the table as things got moved yet again. As I recall, Corel was (and started?? did??) compiling the whole office suite to run with Wine binaries, so in theory you could either run under Windows or under Wine with the same binaries.
It pretty much failed. Was slow and buggy by all accounts. They ran their own custom Wine version. Though it would run side-by-side with an official Wine install. They had the suite (ver. 9 or 10, as well as their PhotoPaint or whatever it was.)
There was also some effort, as I recall, to port portions/the whole mess to Java, but I think that was pretty still born.
FWIW, IMHO, Novell killed a great product and the last real agressive customer sevice organization in consumer computing. The times I called WP help, I may have waited online for some time, but the techs I spoke with _knew_ the product, followed up, and spoke a brand of english I could understand. I was very sorry to see that service go away.
Thanks for the info. Searches for info revealed timetables where the UNIX, VAX, Amiga, and other hardware types were mentioned. There were mentions of their JAVA attempts also in that timeframe. Your recollection helped me piece together some of the events.
I'll lookup Caldera Linux, which was found on some of the searches. (with many different search engines)
Jim
Caldera is now The SCO Group. The Caldera Linux product was OpenLinux. It had a per-seat charge and was targeted at the desktop market. It failed, miserably.
Patrick
On Tue, 2005-27-09 at 06:28 +0200, Fajar Priyanto wrote:
FYI...
http://www.smartofficenews.com.au/Computing/Platforms_And_Applications?artic... e=/Computing/Platforms%20And%20Applications/News/E5T7U6H8&page=1
Is this a threat to Linux?
As most others have said ; No.
Internal disputes are a far greater threat. I am not aware of any significant disputes at the kernel level, but at the distribution level that is another story.
Hard nosed management and developer decisions not supported by the user base may be the downfall. I have been a huge supporter of Linux, and Red Hat since around 1995. Every release of RHL was a tremendous step forward until 8.0 since then there has been one disappointment after another, and I have seen little evidence that things will get better. Even two years ago I would have never entertained the thought of using windows at home. This spring after many heated debates with RH developers about GRUB and LILO, I was still unable to get a stable SATA/IDE mixed machine to run stably. I decided to try XP Pro 2005 this spring and it has been working well. My old machine is now dual boot FC3 and XP Pro as well, and I am using the FC3 less and less. With all the serious problems installing and running FC4, I have not installed it but have been reserving hope that FC5 will be a Saving Grace.
Linux distributions need to get back to their roots if they want to survive. They will hang on lingering with support from only die hard fans, but like me after enough abuse they will start loose interest too. It will take innovation and listening to what the users expect and want, to re-invigorate the user base if Linux distributions are to survive.
--On Tuesday, September 27, 2005 9:57 AM -0600 Guy Fraser guy@incentre.net wrote:
Linux distributions need to get back to their roots if they want to survive. They will hang on lingering with support from only die hard fans, but like me after enough abuse they will start loose interest too. It will take innovation and listening to what the users expect and want, to re-invigorate the user base if Linux distributions are to survive.
You presume that all Linux users are alike, and that all distributions cater to all users. While it's nice to sell Linux to the masses to gain more hardware support, they've never been the core constituency.
On Tuesday 27 September 2005 17:15, Kenneth Porter wrote:
--On Tuesday, September 27, 2005 9:57 AM -0600 Guy Fraser
guy@incentre.net wrote:
Linux distributions need to get back to their roots if they want to survive. They will hang on lingering with support from only die hard fans, but like me after enough abuse they will start loose interest too. It will take innovation and listening to what the users expect and want, to re-invigorate the user base if Linux distributions are to survive.
You presume that all Linux users are alike, and that all distributions cater to all users. While it's nice to sell Linux to the masses to gain more hardware support, they've never been the core constituency.
The major differences may not be visable from a gui like Gnome or KDE, but they are very apparent for those who must administer complex systems. In fact this is where Linux has had the greatest impact.
Control of the system is complete with Linux, unlike MS systems, and this is not apparent unless you worked with these systems where unusual issues, or problems have arisen. On a Linux system terminal you can directly work on the system and programs as root without any controls form the operating system.
Using the configuration gui's will allow a new or unskilled user to effectively manage a typical system. Gui's also allows for faster administration, but, when a system is used for heavier technical use such as a server, gateway, firewall, etc., the need for complete -> direct <- control makes linux a much more flexable and viable system to use.
As far as "die hard fans" are concerned the issue is more ease of use. Those who are skilled, myself included, actually find Linux easier to use and administer.
Another issue is the rapidity in which patches (especially security patches) are available. For any secured system this is a major issue.
For get "die hard fans" it is the most practical system available.
jludwig wrote:
On Tuesday 27 September 2005 17:15, Kenneth Porter wrote:
--On Tuesday, September 27, 2005 9:57 AM -0600 Guy Fraser
guy@incentre.net wrote:
Linux distributions need to get back to their roots if they want to survive. They will hang on lingering with support from only die hard fans, but like me after enough abuse they will start loose interest too. It will take innovation and listening to what the users expect and want, to re-invigorate the user base if Linux distributions are to survive.
You presume that all Linux users are alike, and that all distributions cater to all users. While it's nice to sell Linux to the masses to gain more hardware support, they've never been the core constituency.
The major differences may not be visable from a gui like Gnome or KDE, but they are very apparent for those who must administer complex systems. In fact this is where Linux has had the greatest impact.
Control of the system is complete with Linux, unlike MS systems, and this is not apparent unless you worked with these systems where unusual issues, or problems have arisen. On a Linux system terminal you can directly work on the system and programs as root without any controls form the operating system.
Using the configuration gui's will allow a new or unskilled user to effectively manage a typical system. Gui's also allows for faster administration, but, when a system is used for heavier technical use such as a server, gateway, firewall, etc., the need for complete -> direct <- control makes linux a much more flexable and viable system to use.
As far as "die hard fans" are concerned the issue is more ease of use. Those who are skilled, myself included, actually find Linux easier to use and administer.
Another issue is the rapidity in which patches (especially security patches) are available. For any secured system this is a major issue.
For get "die hard fans" it is the most practical system available.
The news sounds good that Microsoft will try to produce a managable operating system. Some of the points that Linux would be endangered by allowing severe bloat, excessive complexity within the OS are also true factors.
I used to be a "die hard" windows fan when W95 was first released and even bought MSOffice. After running W95 and RHL 5.2 for awhile and becoming exposed with the strengths and the weaknesses of both types of operating systems, I became more of a Linux "die hard". The influencing factors that tilted the tables toward Linux (RHL 5.2) was the support for open source from WordPerfect, RealAudio and Netscape 4.x series browsers. Windows 95 sort of leveled itself to a lower platform by its inferior features of auto-configuring all of the hardware settings from working settings to settings which did not work to make room for new hardware. What happened in my case was the system modem and the soundcard worked. After adding a network card, IRQs and other settings were changed which broke the modem and soundcard. Another factor was setting up RHL 5.2 just worked for getting connected to the Internet. Windows 95 did not even install TCP/IP when a network card was installed. This led me to calling my ISP and complaining that I could not get windows to connect. I could however connect with Linux without any problems. Of course, the phone support staed that Linux was unsupported anyway. I replied again that it was Windows that would not work and got the phone support. Thereafter, I preferred Linux over Windows for most things. I had to dual boot for things Linux would not do for some time. (CD burning, cameras and such). Now, I look at Windows as something that is force sold for most computers. You pay for it anyway, why not use it to fill in the dead spots still left in Linux.
Windows was broken for a long time. They have the resources to make an OS that works. I do not think that Linux is endangered yet. Maybe from some programs that appeal to different distributions, but other programs that actually work are not used instead. I see the sway away from Distro specific tools for Linux distros. Though I am not even going to anticipate how successful or how much of a failure any particular OS will be in the future.
Jim
On Tue, 2005-27-09 at 20:46 -0400, Jim Cornette wrote:
jludwig wrote:
On Tuesday 27 September 2005 17:15, Kenneth Porter wrote:
--On Tuesday, September 27, 2005 9:57 AM -0600 Guy Fraser
guy@incentre.net wrote:
Linux distributions need to get back to their roots if they want to survive. They will hang on lingering with support from only die hard fans, but like me after enough abuse they will start loose interest too. It will take innovation and listening to what the users expect and want, to re-invigorate the user base if Linux distributions are to survive.
You presume that all Linux users are alike, and that all distributions cater to all users. While it's nice to sell Linux to the masses to gain more hardware support, they've never been the core constituency.
...snip...
Windows was broken for a long time. They have the resources to make an OS that works. I do not think that Linux is endangered yet. Maybe from some programs that appeal to different distributions, but other programs that actually work are not used instead. I see the sway away from Distro specific tools for Linux distros. Though I am not even going to anticipate how successful or how much of a failure any particular OS will be in the future.
Jim
Indeed part of what I was saying, was that apparently bureaucratic decisions to remove good software, and replace it with something I was unfamiliar with or was dysfunctional for my equipment or circumstances.
The posts about the GUI admin, are not relevant to my post or situation. I rarely use many of the GUI admin tools, partly because they have bitten me before, but during setup after new installs I try them out to see how they may have improved before using vi if nedit is not yet installed. I have yet to throw any significant support for Windows as a server. The excuses that RH is not interested in desktop systems, but rather focuses on enterprises rings hollow. Many small to medium sized businesses, have some services running from desktop systems that are regularly used as a workstation. The excuse that linux is safe because it has a robust command line interface that is good for remote administration and is reliable does not hold water either. There are other very good and reliable Free Unix server operating systems. The main advantage linux has had over other Free Unix operating systems has been support for desktop hardware and good desktop applications, but that is diminishing at a rapid pace.
The other part I may not have mentioned, was that MS has taken a few lumps and seems to be willing to change things now. If they can develop a good server OS that supports SSH command line administration, they may very well regain significant market share in the server domain and possibly even at the enterprise level. Dumping on administrators that have been very supportive in the past, because the bureaucrats decided to put all the coal in one oven, does not benefit any distro, or the Linux community as a whole.
--On Tuesday, September 27, 2005 6:35 PM -0400 jludwig wralphie@comcast.net wrote:
Control of the system is complete with Linux, unlike MS systems, and this is not apparent unless you worked with these systems where unusual issues, or problems have arisen. On a Linux system terminal you can directly work on the system and programs as root without any controls form the operating system.
Another issue is visibility. When something isn't working right, I can see all the code down to the bare hardware. (I'm of course excluding proprietary drivers such as nVidia's.) I can identify problems myself at the source level and the developers have been great about fixing the issue or accepting a patch and getting me up and running quickly.
Right now I'm coding in the MS world and I curse every time I run into the "source wall" of the OS or a proprietary driver for which I have no source, and can't do more than try to create a reproducible crash.
On Wednesday 28 September 2005 03:49, Kenneth Porter wrote:
Right now I'm coding in the MS world and I curse every time I run into the "source wall" of the OS or a proprietary driver for which I have no source, and can't do more than try to create a reproducible crash.
This I recognise very well and even if it is so my management still doesn't believe that I can't do anything to improve the situation in a short time. I totally depend on suppliers to solve the issue in their driver/firmware. On the other hand I also spend much time in helping out my suppliers. In most cases if their support is not in time or not sufficient it will be the last time I use them as a supplier. So far most support has been adequate, except for MS (who's support was the worst I've ever met). If all my suppliers (some of them do) support Linux, the choice would be very simple for me and I'd move all my things to Linux immediately.
Regards, Marcel
--- Kenneth Porter shiva@sewingwitch.com wrote:
--On Tuesday, September 27, 2005 9:57 AM -0600 Guy Fraser guy@incentre.net wrote:
Linux distributions need to get back to their roots if they want to survive. They will hang on lingering with support from only die hard fans, but like me after enough abuse they will start loose interest too. It will take innovation and listening to what the users expect and want, to re-invigorate the user base if Linux distributions are to survive.
You presume that all Linux users are alike, and that all distributions cater to all users. While it's nice to sell Linux to the masses to gain more hardware support, they've never been the core constituency.
--
And that is a main problem if the goal is to get the masses to seriously consider Linux. If not, then it is a 'geek OS' for 'geeks Only' who are not motivated to make the OS user friendly beyond a centain point.
I have seen this type of environment before at two different (now defunct) Mini-Computer Vendors (remember the 80s?). The Developers tend to think that thier world viewpoint is good enough for everyone. It was always a battle for Tech. Support to get them to see it the customer's way. After all, they were buying the hardware and software.
On Wed, 2005-09-28 at 09:25, BRUCE STANLEY wrote:
And that is a main problem if the goal is to get the masses to seriously consider Linux. If not, then it is a 'geek OS' for 'geeks Only' who are not motivated to make the OS user friendly beyond a centain point.
User-friendly often means hiding the details. There are some Linux distributions that lean this direction. SME server on the server side, xandros, linspire, etc. for desktops. They don't seem to be displacing the more general distributions - and they can be frustrating if you want to do something that the interface designer didn't expect.
I have seen this type of environment before at two different (now defunct) Mini-Computer Vendors (remember the 80s?). The Developers tend to think that thier world viewpoint is good enough for everyone. It was always a battle for Tech. Support to get them to see it the customer's way. After all, they were buying the hardware and software.
That doesn't apply the same way to free software. The developers are the ones who understand the way software should work. Why should they hide options for the people who don't understand if they aren't forced to? On the other hand, good defaults make sense for everyone.
On Wed, 28 Sep 2005, Les Mikesell wrote:
On Wed, 2005-09-28 at 09:25, BRUCE STANLEY wrote:
And that is a main problem if the goal is to get the masses to seriously consider Linux. If not, then it is a 'geek OS' for 'geeks Only' who are not motivated to make the OS user friendly beyond a centain point.
User-friendly often means hiding the details. There are some Linux
Of course it doesn't have to. It can mean that seeing them is usually unnecessary.
I have seen this type of environment before at two different (now defunct) Mini-Computer Vendors (remember the 80s?). The Developers tend to think that thier world viewpoint is good enough for everyone. It was always a battle for Tech. Support to get them to see it the customer's way. After all, they were buying the hardware and software.
That doesn't apply the same way to free software. The developers are the ones who understand the way software should work. Why should they
There is no way but the one true way and developers are its prophets. It should have been easy for me to get rid of vim's awful colors and its automatic indentation. That said, too much automation can be a security hazard.
hide options for the people who don't understand if they aren't forced to? On the other hand, good defaults make sense for everyone.
Yup. That is how one makes seeing the the details usually unnecessary. Of course what's good for a power developer is not necessarily good for a novice. Having flags like --novice and --power-developer that affect multiple other flags can help with this. To assist in the transistion from novice, it shouldn't be too hard to discover what flags are affected by --novice and better yet, why.
On Wed, 2005-09-28 at 12:12, Michael Hennebry wrote:
And that is a main problem if the goal is to get the masses to seriously consider Linux. If not, then it is a 'geek OS' for 'geeks Only' who are not motivated to make the OS user friendly beyond a centain point.
User-friendly often means hiding the details. There are some Linux
Of course it doesn't have to.
Examples would be helpful - perhaps something that makes simple things simple without making difficult things impossible (the normal GUI curse).
It can mean that seeing them is usually unnecessary.
"Usually" doesn't count if you sometimes need them and they have been hidden or removed.
I have seen this type of environment before at two different (now defunct) Mini-Computer Vendors (remember the 80s?). The Developers tend to think that thier world viewpoint is good enough for everyone. It was always a battle for Tech. Support to get them to see it the customer's way. After all, they were buying the hardware and software.
That doesn't apply the same way to free software. The developers are the ones who understand the way software should work. Why should they
There is no way but the one true way and developers are its prophets.
Not quite the point - users may sometimes know what they want done today, but seldom know all the possibilities or permutations of how it should be done or what other users of the same program want. A user would be happy with an ornate piece of precast concrete that just fits the spot he has for it and nowhere else. The developer wants to make bricks that can be laid without having to know all of the possible building shapes ahead of time.
It should have been easy for me to get rid of vim's awful colors and its automatic indentation.
If you don't like embellishments, use vi instead of vim.
That said, too much automation can be a security hazard.
Defaults don't fit everyone. That's why the fine tuning is complicated because the many options are necessary.
hide options for the people who don't understand if they aren't forced to? On the other hand, good defaults make sense for everyone.
Yup. That is how one makes seeing the the details usually unnecessary. Of course what's good for a power developer is not necessarily good for a novice. Having flags like --novice and --power-developer that affect multiple other flags can help with this. To assist in the transistion from novice, it shouldn't be too hard to discover what flags are affected by --novice and better yet, why.
Personally, I think there should be somewhere between 20 and a hundred different distributions that differ only in which programs are installed and their default configurations. Unix historically was developed as a multiuser system with the premise that a skilled administrator would do the local installation and configuration for a set of users. This doesn't mesh well with personal computers and users that install things themselves. One size doesn't quite fit all, but on the other hand every user can't hire a personal system administrator and probably doesn't want to read thousands of pages of manuals to get the background color and fonts he likes. We need a middle ground where a few people configure machines tuned for different uses and preferences and a way for anyone who needs that configuration to copy it. That way, you never really need to know the details of how to do the setup, just where to find one like it. And if you want something that no one else has done - well that's a hint that it's probably a bad idea, but if you make it work you could help everyone else out by sharing it.
On Wed, 28 Sep 2005, Les Mikesell wrote:
On Wed, 2005-09-28 at 12:12, Michael Hennebry wrote:
And that is a main problem if the goal is to get the masses to seriously consider Linux. If not, then it is a 'geek OS' for 'geeks Only' who are not motivated to make the OS user friendly beyond a centain point.
User-friendly often means hiding the details. There are some Linux
Of course it doesn't have to.
Examples would be helpful - perhaps something that makes simple things simple without making difficult things impossible (the normal GUI curse).
It can mean that seeing them is usually unnecessary.
"Usually" doesn't count if you sometimes need them and they have been hidden or removed.
I should have been more clear. "noticing them" instead of "seeing them" is clearer.
The gcc code-generation option that I directly use are -O, -O1, -O2, and -O3.
gcc has a lot of other code generation options that I see, but mostly don't notice.
Other people notice them and use them.
I share your unfondness for a lot of GUI's. A CLI is inherently easier to document with text. (duh) Also a GUI tends to be a moving target, thereby making what documentation there is out of date.
I have seen this type of environment before at two different (now defunct) Mini-Computer Vendors (remember the 80s?). The Developers tend to think that thier world viewpoint is good enough for everyone. It was always a battle for Tech. Support to get them to see it the customer's way. After all, they were buying the hardware and software.
That doesn't apply the same way to free software. The developers are the ones who understand the way software should work. Why should they
There is no way but the one true way and developers are its prophets.
Not quite the point - users may sometimes know what they want done today, but seldom know all the possibilities or permutations of how it should be done or what other users of the same program want. A user would be happy with an ornate piece of precast concrete that just fits the spot he has for it and nowhere else. The developer wants to make bricks that can be laid without having to know all of the possible building shapes ahead of time.
It should have been easy for me to get rid of vim's awful colors and its automatic indentation.
If you don't like embellishments, use vi instead of vim.
Not nice. It tells me that you don't want me to ever get good colors. Though I'm not one of them, there are probably people who like the idea of colored text, but would rather not have vim's garish default combination.
That said, too much automation can be a security hazard.
Defaults don't fit everyone. That's why the fine tuning is complicated because the many options are necessary.
hide options for the people who don't understand if they aren't forced to? On the other hand, good defaults make sense for everyone.
Yup. That is how one makes seeing the the details usually unnecessary. Of course what's good for a power developer is not necessarily good for a novice. Having flags like --novice and --power-developer that affect multiple other flags can help with this. To assist in the transistion from novice, it shouldn't be too hard to discover what flags are affected by --novice and better yet, why.
Personally, I think there should be somewhere between 20 and a hundred different distributions that differ only in which programs are installed and their default configurations. Unix historically was developed as a multiuser system with the premise that a skilled administrator would do the local installation and configuration for a set of users. This doesn't mesh well with personal computers and users that install things themselves. ...
True.
... One size doesn't quite fit all, but on the other hand every user can't hire a personal system administrator and probably doesn't want to read thousands of pages of manuals to get the background color and fonts he likes. We need a middle ground where a few people configure machines tuned for different uses and preferences and a way for anyone who needs that configuration to copy it. That way, you never really need to know the details of how to do the setup, just where to find one like it. And if you want something that no one else has done - well that's a hint that it's probably a bad idea, but if you make it work
Or it's just a combination no distributor thought enough people would want. Who new someone would want Citizen Kane and Debbie Does Dallas or would want whipped cream on his pickle?
On the other hand displaying MPEG-2 in software on a bitty-box is probably a non-starter for anyone.
It might be good to have a place for common software that didn't have to be owned by root. Major mistakes could be "fixed" by wiping out the area and starting over. Not being in root mode would bound the possible damage. If he has been good about backups he wouldn't have to do too much reinstalling.
you could help everyone else out by sharing it.
On Wed, 2005-09-28 at 14:39 -0500, Michael Hennebry wrote:
Also a GUI tends to be a moving target, thereby making what documentation there is out of date.
A good GUI shouldn't need documentation, though; it should explain itself intuitively, and provide some hints for the more difficult bits.
A GUI does make it easier to set either/or options (won't let you do both), and provide other multiple choice options. But some are just woeful, take Evolution's "automatically synchronise remote mail locally" option. What does that mean? The manual doesn't explain. I'd have to do a series of tests to work out if that means keep messages available locally and remotely, copy missing messages or delete local copy of a remote message that was deleted by something else, and so on. I have another mailer on Windows, The Bat!, with a really weird and hazardous sounding option that I've never ticked because I can't find an explanation: "purge unread messages". Why would I want to lose a message I haven't read yet?
On Thu, Sep 29, 2005 at 10:57:03PM +0930, Tim wrote:
On Wed, 2005-09-28 at 14:39 -0500, Michael Hennebry wrote:
Also a GUI tends to be a moving target, thereby making what documentation there is out of date.
A good GUI shouldn't need documentation, though; it should explain itself intuitively, and provide some hints for the more difficult bits.
I have an ex-student who made a claim like this recently. His company produces a product that needs no documentation. It is "intuitively obvious" he says. Balderdash. I am still waiting for the program that needs no documentation. I think I will die first. Linux Journal put me on to search tools recently. I am frustrated that kfind for example has no man page, and I would still like to know how you search jpegs using it. That is certainly not obvious to me and that annoys me.
A GUI does make it easier to set either/or options (won't let you do both), and provide other multiple choice options. But some are just woeful, take Evolution's "automatically synchronise remote mail locally" option. What does that mean? The manual doesn't explain. I'd have to do a series of tests to work out if that means keep messages available locally and remotely, copy missing messages or delete local copy of a remote message that was deleted by something else, and so on. I have another mailer on Windows, The Bat!, with a really weird and hazardous sounding option that I've never ticked because I can't find an explanation: "purge unread messages". Why would I want to lose a message I haven't read yet?
-- Don't send private replies to my address, the mailbox is ignored. I read messages from the public lists.
-- fedora-list mailing list fedora-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
akonstam@trinity.edu wrote:
On Thu, Sep 29, 2005 at 10:57:03PM +0930, Tim wrote:
On Wed, 2005-09-28 at 14:39 -0500, Michael Hennebry wrote:
Also a GUI tends to be a moving target, thereby making what documentation there is out of date.
A good GUI shouldn't need documentation, though; it should explain itself intuitively, and provide some hints for the more difficult bits.
I have an ex-student who made a claim like this recently. His company produces a product that needs no documentation. It is "intuitively obvious" he says. Balderdash. I am still waiting for the program that needs no documentation. I think I will die first. Linux Journal put me on to
My code doesn't need documentation... It's self-documenting. See how obvious it is? And NO COMMENTS!
:-)
We've all heard that line before in a dozen different ways.
[snip]
Mike
On Thu, 2005-29-09 at 17:03 -0500, Mike McCarty wrote:
akonstam@trinity.edu wrote:
On Thu, Sep 29, 2005 at 10:57:03PM +0930, Tim wrote:
On Wed, 2005-09-28 at 14:39 -0500, Michael Hennebry wrote:
Also a GUI tends to be a moving target, thereby making what documentation there is out of date.
A good GUI shouldn't need documentation, though; it should explain itself intuitively, and provide some hints for the more difficult bits.
I have an ex-student who made a claim like this recently. His company produces a product that needs no documentation. It is "intuitively obvious" he says. Balderdash. I am still waiting for the program that needs no documentation. I think I will die first. Linux Journal put me on to
My code doesn't need documentation... It's self-documenting. See how obvious it is? And NO COMMENTS!
:-)
We've all heard that line before in a dozen different ways.
[snip]
Yes, and any of us who have had the misfortune of fixing or modifying such code, know just how wrong they are.
<deleted> How to write legible code with comments and meaningful errors. </deleted>
Tim:
A good GUI shouldn't need documentation, though; it should explain itself intuitively, and provide some hints for the more difficult bits.
akonstam@trinity.edu
I have an ex-student who made a claim like this recently. His company produces a product that needs no documentation. It is "intuitively obvious" he says. Balderdash. I am still waiting for the program that needs no documentation.
Then it's not a "good GUI". :-\ I did say a good GUI shouldn't need it, I didn't say all GUIs are good. A lot are quite crap, like the two examples I gave.
On the other side of the coin, a lot of non-GUI programs are crap to use for similar reasons: Unintuitive ways of working, requires documentation to understand how to use them, and the documentation is poor.
Tim wrote:
Tim:
A good GUI shouldn't need documentation, though; it should explain itself intuitively, and provide some hints for the more difficult bits.
akonstam@trinity.edu
I have an ex-student who made a claim like this recently. His company produces a product that needs no documentation. It is "intuitively obvious" he says. Balderdash. I am still waiting for the program that needs no documentation.
Then it's not a "good GUI". :-\ I did say a good GUI shouldn't need it, I didn't say all GUIs are good. A lot are quite crap, like the two examples I gave.
ANY significant system needs documentation. I take it that the only good GUI is a trivial, useless one, then.
On the other side of the coin, a lot of non-GUI programs are crap to use for similar reasons: Unintuitive ways of working, requires documentation to understand how to use them, and the documentation is poor.
Nobody has said otherwise, AFAIK.
Mike
Mike McCarty wrote:
Tim wrote:
Tim:
A good GUI shouldn't need documentation, though; it should explain itself intuitively, and provide some hints for the more difficult bits.
akonstam@trinity.edu
I have an ex-student who made a claim like this recently. His company produces a product that needs no documentation. It is "intuitively obvious" he says. Balderdash. I am still waiting for the program that needs no documentation.
Then it's not a "good GUI". :-\ I did say a good GUI shouldn't need it, I didn't say all GUIs are good. A lot are quite crap, like the two examples I gave.
ANY significant system needs documentation. I take it that the only good GUI is a trivial, useless one, then.
On the other side of the coin, a lot of non-GUI programs are crap to use for similar reasons: Unintuitive ways of working, requires documentation to understand how to use them, and the documentation is poor.
Nobody has said otherwise, AFAIK.
Mike
You can't make a one size fits all gui. What's intuitive to one person might be obscure to someone else. Everybody's expectations are different, where you would expect to find Preferences... (assuming they are even called that) might be the last place I'd look. Much as I hate to say it, that's one thing we have to thank Microsoft for. Windows has moulded the expectations of the great unwashed to understand "intuitively" how to work PC software. That means the need for documentation has perhaps reduced but certainly not eliminated. If a piece of software I install doesn't have a help file, it's instant deinstall and no correspondence will be entered into.
--On Wednesday, September 28, 2005 12:12 PM -0500 Michael Hennebry hennebry@web.cs.ndsu.nodak.edu wrote:
There is no way but the one true way and developers are its prophets. It should have been easy for me to get rid of vim's awful colors and its automatic indentation.
There's an easy fix for that: Code the change yourself, or pay someone to do it. Free software doesn't mean everyone else has to do scratch *your* itch for free. If you want things a certain way, then *make* them that way. That's why we have the GPL. It's to empower you with this capability.
Michael Hennebry wrote:
There is no way but the one true way and developers are its prophets. It should have been easy for me to get rid of vim's awful colors and its automatic indentation. That said, too much automation can be a security hazard.
Hang on: you're making some good points about user-friendliness, but you're seriously using *vim* as an example?
Don't get me wrong: it's a great editor, and I'm typing this into it now. But it's *not* aimed at the average user: it's aimed at people who know how vi works or want to learn. That really does include knowing a bit about Unix. In other words, I *do* think it's reasonable for vim's developers to expect users to know that per-user configuration for a program foo might just be held in ~/.foorc .
And are syntax off and set noai *so* difficult to find out about? set [no]ai is there in plain and standard vi, for goodness' sake!
James.
On Thursday 29 September 2005 02:35, James Wilkinson wrote:
Michael Hennebry wrote:
There is no way but the one true way and developers are its prophets. It should have been easy for me to get rid of vim's awful colors and its automatic indentation. That said, too much automation can be a security hazard.
Hang on: you're making some good points about user-friendliness, but you're seriously using *vim* as an example?
Don't get me wrong: it's a great editor, and I'm typing this into it now. But it's *not* aimed at the average user: it's aimed at people who know how vi works or want to learn. That really does include knowing a bit about Unix. In other words, I *do* think it's reasonable for vim's developers to expect users to know that per-user configuration for a program foo might just be held in ~/.foorc .
And are syntax off and set noai *so* difficult to find out about? set [no]ai is there in plain and standard vi, for goodness' sake!
James.
-- E-mail address: james | Five miles as the hippopotamus bounces... @westexe.demon.co.uk |
Which brings another point. You are not limited to a particular program or group of programs S.A Word, Note Pad, et al, built with the same corporate policies and philosophies. Indeed there are usually several divergent options to choose from each with it's own strengths and weaknesses.
On Tue, 2005-09-27 at 09:57 -0600, Guy Fraser wrote:
Every release of RHL was a tremendous step forward until 8.0 since then there has been one
What was wrong?
Linux distributions need to get back to their roots if they want to survive.
What are those roots the Linux went from?
On Fri, 2005-09-30 at 12:22, Strong wrote:
Every release of RHL was a tremendous step forward until 8.0 since then there has been one
What was wrong?
Linux distributions need to get back to their roots if they want to survive.
What are those roots the Linux went from?
Someone who feels this way might like RHEL3 or the free Centos3.x distribution. This is based on the 2.4 kernel and stuff that's been running for years but still getting timely bugfix and security updates. The roots are still around for people who don't like surprises.
On Fri, 2005-30-09 at 13:46 -0500, Les Mikesell wrote:
On Fri, 2005-09-30 at 12:22, Strong wrote:
Every release of RHL was a tremendous step forward until 8.0 since then there has been one
What was wrong?
Linux distributions need to get back to their roots if they want to survive.
What are those roots the Linux went from?
Someone who feels this way might like RHEL3 or the free Centos3.x distribution. This is based on the 2.4 kernel and stuff that's been running for years but still getting timely bugfix and security updates. The roots are still around for people who don't like surprises.
It had nothing to do with the 2.6 kernel or frequent updates. It had to do with dropping support for many of the command line and applications many people used and wanted to continue using. It also involved changes to the kernel that made it impossible for most people to get some of their applications to compile, which is partly why they were dropped by Red Hat. Unfortunately if anyone were to build their own kernel, they would then have to constantly maintain it, whenever security vulnerabilities were discovered. In any case things broke without warning and no remedies were provided by Red Hat, the package was just dropped in the next release. Many of these packages are still maintained and available on other distributions, contrary to many of the excuses given at the time.
I can not continue, I feel a rant coming on.
On Fri, 2005-09-30 at 17:42, Guy Fraser wrote:
Every release of RHL was a tremendous step forward until 8.0 since then there has been one
What was wrong?
It had nothing to do with the 2.6 kernel or frequent updates. It had to do with dropping support for many of the command line and applications many people used and wanted to continue using. It also involved changes to the kernel that made it impossible for most people to get some of their applications to compile, which is partly why they were dropped by Red Hat. Unfortunately if anyone were to build their own kernel, they would then have to constantly maintain it, whenever security vulnerabilities were discovered. In any case things broke without warning and no remedies were provided by Red Hat, the package was just dropped in the next release. Many of these packages are still maintained and available on other distributions, contrary to many of the excuses given at the time.
I can not continue, I feel a rant coming on.
I can't think of anything offhand that worked on RH7.x that doesn't work in Centos3.x. Elm, maybe... Examples???
-- Les Mikesell lesmikesell@gmail.com
On Fri, 2005-30-09 at 21:02 -0500, Les Mikesell wrote:
On Fri, 2005-09-30 at 17:42, Guy Fraser wrote:
Every release of RHL was a tremendous step forward until 8.0 since then there has been one
What was wrong?
It had nothing to do with the 2.6 kernel or frequent updates. It had to do with dropping support for many of the command line and applications many people used and wanted to continue using. It also involved changes to the kernel that made it impossible for most people to get some of their applications to compile, which is partly why they were dropped by Red Hat. Unfortunately if anyone were to build their own kernel, they would then have to constantly maintain it, whenever security vulnerabilities were discovered. In any case things broke without warning and no remedies were provided by Red Hat, the package was just dropped in the next release. Many of these packages are still maintained and available on other distributions, contrary to many of the excuses given at the time.
I can not continue, I feel a rant coming on.
I can't think of anything offhand that worked on RH7.x that doesn't work in Centos3.x. Elm, maybe... Examples???
Elm, Pico, Wine, Word Perfect and Gatos Drivers are all I can think of now. Some of these may have been fixed by now but I have replaced hardware and started using terminal services to use the programs I needed to have running in wine. Some of these are available in other repositories now, but that did not help back then. It also didn't help how I felt when I was billed for a one year RHN subscription a month before the announcement that in 6 months RHL 9 would no longer be supported.
It's a little late to be concerned about this now, the reputation was damaged quite a while ago. And the arrogant attitudes that caused the rift, persist to this day. Unless you have pull within RH, I suggest you drop the inquisition, all you are doing is reminding me and many others of problems that upset them from back around that time. If I didn't have customers using RH products I would not be here now. I have moved all our servers to FreeBSD and only keep one FC workstation around for testing.
On Mon, 2005-10-03 at 10:21 -0600, Guy Fraser wrote:
On Fri, 2005-30-09 at 21:02 -0500, Les Mikesell wrote:
On Fri, 2005-09-30 at 17:42, Guy Fraser wrote:
Every release of RHL was a tremendous step forward until 8.0 since then there has been one
What was wrong?
It had nothing to do with the 2.6 kernel or frequent updates. It had to do with dropping support for many of the command line and applications many people used and wanted to continue using. It also involved changes to the kernel that made it impossible for most people to get some of their applications to compile, which is partly why they were dropped by Red Hat. Unfortunately if anyone were to build their own kernel, they would then have to constantly maintain it, whenever security vulnerabilities were discovered. In any case things broke without warning and no remedies were provided by Red Hat, the package was just dropped in the next release. Many of these packages are still maintained and available on other distributions, contrary to many of the excuses given at the time.
I can not continue, I feel a rant coming on.
I can't think of anything offhand that worked on RH7.x that doesn't work in Centos3.x. Elm, maybe... Examples???
Elm, Pico, Wine, Word Perfect and Gatos Drivers are all I can think of now. Some of these may have been fixed by now but I have replaced hardware and started using terminal services to use the programs I needed to have running in wine. Some of these are available in other repositories now, but that did not help back then. It also didn't help how I felt when I was billed for a one year RHN subscription a month before the announcement that in 6 months RHL 9 would no longer be supported.
---- boy - you sure take $ 60 to the max. I think in all, the sweeping changes of dropping RHL and subscriptions for RHL in favor of Fedora allowed them to concentrate on the one revenue product - RHEL and this has been good for all except those that use Fedora because it was free. CentOS gives you the free version of RHEL which totally makes sense and is a great ally to RHEL/Fedora. ----
It's a little late to be concerned about this now, the reputation was damaged quite a while ago. And the arrogant attitudes that caused the rift, persist to this day. Unless you have pull within RH, I suggest you drop the inquisition, all you are doing is reminding me and many others of problems that upset them from back around that time. If I didn't have customers using RH products I would not be here now. I have moved all our servers to FreeBSD and only keep one FC workstation around for testing.
---- FreeBSD is a great option too.
Fedora seems to be the cutting edge for desktops at the moment but I wouldn't want to run a server on it.
Craig
On Mon, 2005-03-10 at 09:36 -0700, Craig White wrote:
On Mon, 2005-10-03 at 10:21 -0600, Guy Fraser wrote:
On Fri, 2005-30-09 at 21:02 -0500, Les Mikesell wrote:
On Fri, 2005-09-30 at 17:42, Guy Fraser wrote:
> Every release of RHL was a > tremendous step forward until 8.0 since then there has been one What was wrong?
It had nothing to do with the 2.6 kernel or frequent updates. It had to do with dropping support for many of the command line and applications many people used and wanted to continue using. It also involved changes to the kernel that made it impossible for most people to get some of their applications to compile, which is partly why they were dropped by Red Hat. Unfortunately if anyone were to build their own kernel, they would then have to constantly maintain it, whenever security vulnerabilities were discovered. In any case things broke without warning and no remedies were provided by Red Hat, the package was just dropped in the next release. Many of these packages are still maintained and available on other distributions, contrary to many of the excuses given at the time.
I can not continue, I feel a rant coming on.
I can't think of anything offhand that worked on RH7.x that doesn't work in Centos3.x. Elm, maybe... Examples???
Elm, Pico, Wine, Word Perfect and Gatos Drivers are all I can think of now. Some of these may have been fixed by now but I have replaced hardware and started using terminal services to use the programs I needed to have running in wine. Some of these are available in other repositories now, but that did not help back then. It also didn't help how I felt when I was billed for a one year RHN subscription a month before the announcement that in 6 months RHL 9 would no longer be supported.
boy - you sure take $ 60 to the max. I think in all, the sweeping changes of dropping RHL and subscriptions for RHL in favor of Fedora allowed them to concentrate on the one revenue product - RHEL and this has been good for all except those that use Fedora because it was free. CentOS gives you the free version of RHEL which totally makes sense and is a great ally to RHEL/Fedora.
It was not so much the cost, but the perceived deceit of taking the money for something they were not intending on providing. I am VERY happy with my decision. We have had our own local FreeBSD repository that is rsynced every night ever since we switched, so we have only had to incur the cost of downloading all the source and updates once. We now have much faster access to all the files required to build new servers and update existing ones. There haven't been any big surprises when applying updates, and the servers have been secure and more reliable, since the switch. Their is also a lot more software available in the ports system repository than in any of the RH distributions and the Extras combined.
It's a little late to be concerned about this now, the reputation was damaged quite a while ago. And the arrogant attitudes that caused the rift, persist to this day. Unless you have pull within RH, I suggest you drop the inquisition, all you are doing is reminding me and many others of problems that upset them from back around that time. If I didn't have customers using RH products I would not be here now. I have moved all our servers to FreeBSD and only keep one FC workstation around for testing.
FreeBSD is a great option too.
Fedora seems to be the cutting edge for desktops at the moment but I wouldn't want to run a server on it. From problems I have had supporting RHEL 4 for customers, I would not
use it either. RH has failed to help the customers in every instance before I was able to fix the problems. They had all bought into the Commercially Supported Product stuff when making the decision, over other distributions and FreeBSD.
On Mon, 2005-10-03 at 11:10 -0600, Guy Fraser wrote:
On Mon, 2005-03-10 at 09:36 -0700, Craig White wrote:
On Mon, 2005-10-03 at 10:21 -0600, Guy Fraser wrote:
On Fri, 2005-30-09 at 21:02 -0500, Les Mikesell wrote:
On Fri, 2005-09-30 at 17:42, Guy Fraser wrote:
> > Every release of RHL was a > > tremendous step forward until 8.0 since then there has been one > What was wrong? >
It had nothing to do with the 2.6 kernel or frequent updates. It had to do with dropping support for many of the command line and applications many people used and wanted to continue using. It also involved changes to the kernel that made it impossible for most people to get some of their applications to compile, which is partly why they were dropped by Red Hat. Unfortunately if anyone were to build their own kernel, they would then have to constantly maintain it, whenever security vulnerabilities were discovered. In any case things broke without warning and no remedies were provided by Red Hat, the package was just dropped in the next release. Many of these packages are still maintained and available on other distributions, contrary to many of the excuses given at the time.
I can not continue, I feel a rant coming on.
I can't think of anything offhand that worked on RH7.x that doesn't work in Centos3.x. Elm, maybe... Examples???
Elm, Pico, Wine, Word Perfect and Gatos Drivers are all I can think of now. Some of these may have been fixed by now but I have replaced hardware and started using terminal services to use the programs I needed to have running in wine. Some of these are available in other repositories now, but that did not help back then. It also didn't help how I felt when I was billed for a one year RHN subscription a month before the announcement that in 6 months RHL 9 would no longer be supported.
boy - you sure take $ 60 to the max. I think in all, the sweeping changes of dropping RHL and subscriptions for RHL in favor of Fedora allowed them to concentrate on the one revenue product - RHEL and this has been good for all except those that use Fedora because it was free. CentOS gives you the free version of RHEL which totally makes sense and is a great ally to RHEL/Fedora.
It was not so much the cost, but the perceived deceit of taking the money for something they were not intending on providing.
---- I don't believe that they had knowledge of these plans until just shortly before their announcement. You obviously give them benefit of no doubts. I can't help that.
I too had paid my RHL subscription and found it relatively worthless after their announcement to change things but I didn't take the same way you did. ----
It's a little late to be concerned about this now, the reputation was damaged quite a while ago. And the arrogant attitudes that caused the rift, persist to this day. Unless you have pull within RH, I suggest you drop the inquisition, all you are doing is reminding me and many others of problems that upset them from back around that time. If I didn't have customers using RH products I would not be here now. I have moved all our servers to FreeBSD and only keep one FC workstation around for testing.
FreeBSD is a great option too.
Fedora seems to be the cutting edge for desktops at the moment but I wouldn't want to run a server on it. From problems I have had supporting RHEL 4 for customers, I would not
use it either. RH has failed to help the customers in every instance before I was able to fix the problems. They had all bought into the Commercially Supported Product stuff when making the decision, over other distributions and FreeBSD.
---- since this is Fedora list - this all is pretty much off-topic
there's no substitute for a good sysadmin. Obviously support from any organization can handle the low hanging fruit of problem resolution on any package.
The issue becomes the more obtuse problems.
With RHEL, you have more than support though, you have certification (both hardware and software).
Craig
On Mon, 2005-10-03 at 11:21, Guy Fraser wrote:
I can't think of anything offhand that worked on RH7.x that doesn't work in Centos3.x. Elm, maybe... Examples???
Elm, Pico, Wine, Word Perfect and Gatos Drivers are all I can think of now. Some of these may have been fixed by now but I have replaced hardware and started using terminal services to use the programs I needed to have running in wine. Some of these are available in other repositories now, but that did not help back then. It also didn't help how I felt when I was billed for a one year RHN subscription a month before the announcement that in 6 months RHL 9 would no longer be supported.
If you resent paying RH for support, you should be really happy to find the Centos distribution...
It's a little late to be concerned about this now, the reputation was damaged quite a while ago. And the arrogant attitudes that caused the rift, persist to this day. Unless you have pull within RH, I suggest you drop the inquisition, all you are doing is reminding me and many others of problems that upset them from back around that time.
Again, Centos does not inherit any attitude problems.
If I didn't have customers using RH products I would not be here now. I have moved all our servers to FreeBSD and only keep one FC workstation around for testing.
That's actually pretty funny if you look at the number of things that worked well on RH9 but not RHEL3 compared to FreeBSD back then.
Les Mikesell wrote:
On Mon, 2005-10-03 at 11:21, Guy Fraser wrote:
[snip]
If you resent paying RH for support, you should be really happy to find the Centos distribution...
Or Scientific Linux...
It's a little late to be concerned about this now, the reputation was damaged quite a while ago. And the arrogant attitudes that caused the rift, persist to this day. Unless you have pull within RH, I suggest you drop the inquisition, all you are doing is reminding me and many others of problems that upset them from back around that time.
Again, Centos does not inherit any attitude problems.
Umm, not quite. Some of the "upgrade or lose support" pressure filters down, due to the fact that it is derived from RHEL. But I agree that it is much better in this respect.
[snip]
Mike
Mike McCarty wrote:
Les Mikesell wrote:
Again, Centos does not inherit any attitude problems.
Umm, not quite. Some of the "upgrade or lose support" pressure filters down, due to the fact that it is derived from RHEL. But I agree that it is much better in this respect.
I really don't know what you're talking about. RHEL has a lifetime of 5 years, and there's little reason to expect that CentOS won't.
On Wed, 2005-10-05 at 12:09, Gordon Messmer wrote:
Again, Centos does not inherit any attitude problems.
Umm, not quite. Some of the "upgrade or lose support" pressure filters down, due to the fact that it is derived from RHEL. But I agree that it is much better in this respect.
I really don't know what you're talking about. RHEL has a lifetime of 5 years, and there's little reason to expect that CentOS won't.
Yes, Centos 3.x is still actively being updated. Besides, even if you do decide to do a version upgrade you can't complain about the price. I'm inclined to think that Linux kernels become mature somewhere in the 2.x.20 range so I'm not in any hurry to switch everything to a 2.6 based distro yet.
On Mon, 2005-10-03 at 12:12 -0500, Les Mikesell wrote:
If you resent paying RH for support, you should be really happy to find the Centos distribution...
Can You explain a bit on this? Why paying for RH support, can I not just download RH system and enjoy using it? Then, why CentOS?!
Strong wrote:
On Mon, 2005-10-03 at 12:12 -0500, Les Mikesell wrote:
If you resent paying RH for support, you should be really happy to find the Centos distribution...
Can You explain a bit on this? Why paying for RH support, can I not just download RH system and enjoy using it? Then, why CentOS?!
I've not seen what went before.
You cannot download the binaries for RHEL. You _can_ download the source and build it yourself if you have a suitable build platform (FC3 is probably good to start with, but build, install the result, build again to be sure) or you can download one of the clones whose providers have done the build-from-source trick.
On Fri, 2005-11-25 at 13:13, Strong wrote:
If you resent paying RH for support, you should be really happy to find the Centos distribution...
Can You explain a bit on this? Why paying for RH support, can I not just download RH system and enjoy using it? Then, why CentOS?!
You can't download the RH enterprise versions without paying the support fee. CentOS is the same code rebuilt from the freely available source rpms minus the trademarked name and artwork. There is also an active mail list for help. http://www.centos.org/
-
akonstam＠trinity.edu -
BRUCE STANLEY -
Craig White -
Fajar Priyanto -
Gordon Messmer -
Guy Fraser -
Ian -
James Wilkinson -
Jeff Vian -
Jim Cornette -
jludwig -
John Summerfield -
Kenneth Porter -
korgull -
Les Mikesell -
Matthew Miller -
Michael A. Peters -
Michael Hennebry -
Mike McCarty -
Patrick May -
Peter Gordon -
Scot L. Harris -
Strong -
thomas cameron -
Tim -
tom