Canonical Saw ~$119M Revenue In 2019 But Still Operating At A Loss

[lyamc]

use a good search engine like duckduckgo and apply this search term "Canonical ubuntu nih syndrome"

I'll humor you.

UEC vs OpenStack

UEC (Ubuntu Enterprise Cloud) was offered by Canonical before Openstack existed. Openstack was released 2010-10-21, Ubuntu incorporated Openstack to replaced their UEC in Ubuntu 11.04 (2011-04). You can't claim NIH when the product didn't exist, and especially can't claim it if Ubuntu switches the moment the better alternative appears.

Bazaar vs Git

Bazaar was released on 2007-12-14 by Canonical while Git was version 1.5. There's a nice little wiki that compares the two, also it makes it clear that Canonical would have been developing Bazaar for at least as long as Git was started since they integrated it with so many things: GUIs, plugin support, native Windows support, etc.

Do you not think that Bazaar might have lead to Git adding more features that Bazaar had? Also do you not believe that due to git, bazaar's performance was significantly improved? Aka, competition is good.

Upstart vs Systemd

Upstart was released August 24, 2006 and systemd was released 30 March 2010. How again is Upstart a NIH Syndrome? God damn you guys are stupid.

Launchpad vs Github

This one is just a repeat of the Bazaar thing. Bazaar integrates with Launchpad, and Launchpad is something that they control. Remember how Github was purchased by Microsoft? Good thing Ubuntu stuck with Launchpad.

Unity vs gnome

Unity IS the gnome shell that uses gnome packages. Unity was only developed because Gnome 3 was coming. Unity was added to Ubuntu in 2011, the same year that Gnome 3 was released. I'd also like to point out that Gnome 3 was terrible for many years: bad performance, stupid interface, missing features. You could claim that Unity 8 was NIH, but no one else had a proper desktop-mobile interface so I would doubt that. The interface and the way to navigate it was totally different.

Torvalds abandoned GNOME for a while after the release of GNOME 3.0, saying "The developers have apparently decided that it's 'too complicated' to actually do real work on your desktop, and have decided to make it really annoying to do".

mir vs wayland

Mir was more than just a display server protocol. The work that went into mir was easily modified to work with Wayland.

It's the most NIH case I can think of, but even that isn't solely a competing protocol and ended up adopting it.

[duby229]

I'll humor you.

UEC vs OpenStack

UEC (Ubuntu Enterprise Cloud) was offered by Canonical before Openstack existed. Openstack was released 2010-10-21, Ubuntu incorporated Openstack to replaced their UEC in Ubuntu 11.04 (2011-04). You can't claim NIH when the product didn't exist, and especially can't claim it if Ubuntu switches the moment the better alternative appears.

Bazaar vs Git

Bazaar was released on 2007-12-14 by Canonical while Git was version 1.5. There's a nice little wiki that compares the two, also it makes it clear that Canonical would have been developing Bazaar for at least as long as Git was started since they integrated it with so many things: GUIs, plugin support, native Windows support, etc.

Do you not think that Bazaar might have lead to Git adding more features that Bazaar had? Also do you not believe that due to git, bazaar's performance was significantly improved? Aka, competition is good.

Upstart vs Systemd

Upstart was released August 24, 2006 and systemd was released 30 March 2010. How again is Upstart a NIH Syndrome? God damn you guys are stupid.

Launchpad vs Github

This one is just a repeat of the Bazaar thing. Bazaar integrates with Launchpad, and Launchpad is something that they control. Remember how Github was purchased by Microsoft? Good thing Ubuntu stuck with Launchpad.

Unity vs gnome

Unity IS the gnome shell that uses gnome packages. Unity was only developed because Gnome 3 was coming. Unity was added to Ubuntu in 2011, the same year that Gnome 3 was released. I'd also like to point out that Gnome 3 was terrible for many years: bad performance, stupid interface, missing features. You could claim that Unity 8 was NIH, but no one else had a proper desktop-mobile interface so I would doubt that. The interface and the way to navigate it was totally different.



mir vs wayland

Mir was more than just a display server protocol. The work that went into mir was easily modified to work with Wayland.

It's the most NIH case I can think of, but even that isn't solely a competing protocol and ended up adopting it.

Thank you. At the very least I can respect your opinion, although I don't share them.

I would like to point out that your examples above are maybe the worst examples of nih at canonical, but they only represent a tiny proportion of the nih they actually wasted.

UEC vs OpenStack is the only example listed above that I don't consider as nih and there are literally dozens of other examples not listed there at all.

[lyamc]

Thank you. At the very least I can respect your opinion, although I don't share them.

I would like to point out that your examples above are maybe the worst examples of nih at canonical, but they only represent a tiny proportion of the nih they actually wasted.

UEC vs OpenStack is the only example listed above that I don't consider as nih and there are literally dozens of other examples not listed there at all.

I listed the most well-known ones, most of which are ironically not even NIH since the thing that it is supposedly re-inventing has not even existed. While I have my opinions listed, it is not an opinion that most of those CAN'T be "not invented here" because they weren't invented.

Also, it's a total cop-out to say "well just search for it" and then I do, tell you they're all bogus with evidence and then you say "well I disagree because there are dozens of worse cases". Really? Maybe YOU should search for some examples before you tell someone else to do it

[jo-erlend]

haha you're scratching the bottom of the barrel. X11 and Xorg are the same, X11 is the protocol and Xorg the software implementing it. And Mutter and Kwin can run on Xorg or Wayland, but not Mir because it sucks and lost. Deal with it even if you're butthurt.

PS: And Mir isn't a replacement for Mutter or KWin - which just shows you have no clue what Mir is.

No, X11 and Xorg are not the same. One is a protocol and the other is a computer program. There are different computer programs using the protocols. For instance, X2go uses X11, but not Xorg. This type of confusion was very common in the 1990s, when a lot of people thought that IRC and mIRC was the same thing. Some people still believe that internet and the web is the same, but it isn't. It's a bit sad to see this type of misunderstanding on a site like Phoronix, where you'd expect people to have some basic knowledge.

Mir is a display server that accepts Wayland and X11 clients and yes, it can be run on Xorg.

The statement that thing run _on_ Wayland, is very strange. Wayland isn't a program. It's a protocol; it defines how computer programs communicate with each other. Mir can easily be compared to Mutter and Kwin as Wayland compositors, even if Mir is designed to be more generic.

[JackLilhammers]

Umm because Canonical has dramatically hurt linux adoption over and over and over and over and over.......................................

That's hardly true. In fact I think it's quite the opposite.
Ubuntu was not the first "easy distro", but sure enough is the one that stood out of the crowd!
Anyway I'm curious about how they have hindered Linux adoption, in your opinion

[jo-erlend]

Whether it's an API or a protocol or you dog on steroids doesn't matter, I said they should've switched to and helped develop Wayland rather than waste resources on Mir - both of which were meant as "the new replacement for X11" - and that's the only point there is. You started the irrelevant non-sense about protocol vs API.

No, Mir replaces Xorg, not X11. It was designed in a way where users should not care which protocols were used internally, whether it was X11, Wayland or some other protocol, because designing public protocols that supports all use-cases is much more difficult than implementing a server and a client. Now Mir is a Wayland compositor like any other, except more advanced in some ways and more generic than most.

The reason you think these differences are irrelevant is because you don't understand. In any other situation, the difference would be completely obvious to you. Like if someone told you that Firefox is the same as HTTP, you would object, wouldn't you? Of course you would. HTTP is a protocol. Firefox is a program. It's not the same thing.

[lyamc]

Say what you want about Canonical, but at least I can buy hardware that has Linux preloaded thanks to them.

[duby229]

I listed the most well-known ones, most of which are ironically not even NIH since the thing that it is supposedly re-inventing has not even existed. While I have my opinions listed, it is not an opinion that most of those CAN'T be "not invented here" because they weren't invented.

Also, it's a total cop-out to say "well just search for it" and then I do, tell you they're all bogus with evidence and then you say "well I disagree because there are dozens of worse cases". Really? Maybe YOU should search for some examples before you tell someone else to do it

Bro, it is -the- simplest kind of internet search possible. If you are so incapable of finding information for yourself, how in the hell is that my fault? Now you're adding your own ignorance as a factor.

You can make the bullshit claim that those examples aren't nih, but the fact is, all but one of them listed above are -THE- worst examples of nih in all of open source.

[duby229]

That's hardly true. In fact I think it's quite the opposite.
Ubuntu was not the first "easy distro", but sure enough is the one that stood out of the crowd!
Anyway I'm curious about how they have hindered Linux adoption, in your opinion

Of course Ubuntu stands out of the crowd, Canonical has committed nih dozens of times and used Ubuntu as the medium. And every single time another one of their nih projects fail it again hurts linux adoption. It's happened dozens of times.

[lyamc]

Bro, it is -the- simplest kind of internet search possible. If you are so incapable of finding information for yourself, how in the hell is that my fault? Now you're adding your own ignorance as a factor.

You can make the bullshit claim that those examples aren't nih, but the fact is, all but one of them listed above are -THE- worst examples of nih in all of open source.

You retard, if it’s so simple to search for, you do it.

Also, I never made the initial claims, you did! You said that Canonical hurts desktop adoption and other claims. You have nothing to back them up.

Upstart was released years before systemd. Tell me precisely how it’s bullshit to claim that there is no evidence of a strong NIH syndrome that cripples linux desktop adoption, or take your bullshit claims elsewhere.