In the early days, many of us wanted Unix at home. I started with a 286 and DOS with various unix utilities that had been ported. I dual booted with Minix. I still couldn't run most GNU software, including Emacs. I purchased a 486 and tried 386BSD but it didn't work. I downloaded SLS Linux with 0.92 kernel. Slackware was created to fix the bugs SLS wasn't fixing, using SLS's package format.
I became a Unix Sysadmin shortly after. I was installing/using Unix on SGI, Sun (Sun3, Sparc, SunOS, Solaris), DEC (Vax, Alpha, MIPS), HP, Apollo, Cray, Intergraph and IIRC Tektronix. Being able to run a near Unix at home w/o purchasing a $20,000 workstation was huge!
The commercial Unixes for PCs (386 & up) were expensive too. $10,000 IIRC. Linux really had no competition. Solaris for x86 wasn't really available until 2.4 in late 1994 and IIRC you still had to purchase media, etc until Solaris 2.6 or maybe later. By that time, Linux was the preferred choice on x86.
The decision to split into Fedora and Redhat Enterprise Linux probably saved the company. Fedora is free and has all the source that potentially goes into RHEL. RHEL is subscription based and companies renew their subscription to get support. Before that, there were some companies that were buying multiple boxed sets just to support Redhat.
One thing people get wrong is that Redhat has stayed Redhat. IBM expects Redhat to meet financial targets but does not mange how. All the decisions are still made by Redhat. When you say IBM made the decision, it did not.
The decision to stop CentOS was communicated poorly, including internally. I don't want to argue the merits of that.
CentOS Stream was created (poor choice of name IMO) to be in between Fedora and RHEL. Fedora is very desktop oriented and its not used for servers often. Companies were not contributing as much to Fedora and CentOS, which is downstream from RHEL was getting nothing. Stream is much closer to RHEL. As a result, companies are contributing patches and updates and they can now influence RHEL. Which is the whole point of opensource.
5 years after IBM purchased Redhat, IBM had made their money back. They lead their financial results with Redhat. As long as Redhat performs, IBM won't change it because that will wreck it. Redhat is still Redhat culturally and is still making its own decisions.
One of the keys to Redhat is that they always contribute to the upstream products. When they buy a company, like Quay for example, they opensouce its code. They're one of the largest contributors to Kubernetes, which was created by Google. They've created or helped create many foundations that help protect opensource against the patent trolls. I don't think any company has done as much for opensource as Redhat.
In the early days, many of us wanted Unix at home. I started with a 286 and DOS with various unix utilities that had been ported. I dual booted with Minix. I still couldn't run most GNU software, including Emacs. I purchased a 486 and tried 386BSD but it didn't work. I downloaded SLS Linux with 0.92 kernel. Slackware was created to fix the bugs SLS wasn't fixing, using SLS's package format.
I became a Unix Sysadmin shortly after. I was installing/using Unix on SGI, Sun (Sun3, Sparc, SunOS, Solaris), DEC (Vax, Alpha, MIPS), HP, Apollo, Cray, Intergraph and IIRC Tektronix. Being able to run a near Unix at home w/o purchasing a $20,000 workstation was huge!
The commercial Unixes for PCs (386 & up) were expensive too. $10,000 IIRC. Linux really had no competition. Solaris for x86 wasn't really available until 2.4 in late 1994 and IIRC you still had to purchase media, etc until Solaris 2.6 or maybe later. By that time, Linux was the preferred choice on x86.
The decision to split into Fedora and Redhat Enterprise Linux probably saved the company. Fedora is free and has all the source that potentially goes into RHEL. RHEL is subscription based and companies renew their subscription to get support. Before that, there were some companies that were buying multiple boxed sets just to support Redhat.
One thing people get wrong is that Redhat has stayed Redhat. IBM expects Redhat to meet financial targets but does not mange how. All the decisions are still made by Redhat. When you say IBM made the decision, it did not.
The decision to stop CentOS was communicated poorly, including internally. I don't want to argue the merits of that.
CentOS Stream was created (poor choice of name IMO) to be in between Fedora and RHEL. Fedora is very desktop oriented and its not used for servers often. Companies were not contributing as much to Fedora and CentOS, which is downstream from RHEL was getting nothing. Stream is much closer to RHEL. As a result, companies are contributing patches and updates and they can now influence RHEL. Which is the whole point of opensource.
5 years after IBM purchased Redhat, IBM had made their money back. They lead their financial results with Redhat. As long as Redhat performs, IBM won't change it because that will wreck it. Redhat is still Redhat culturally and is still making its own decisions.
One of the keys to Redhat is that they always contribute to the upstream products. When they buy a company, like Quay for example, they opensouce its code. They're one of the largest contributors to Kubernetes, which was created by Google. They've created or helped create many foundations that help protect opensource against the patent trolls. I don't think any company has done as much for opensource as Redhat.