GNOME Gets New Versioning Scheme (betanews.com) 91
AmiMoJo writes: The GNOME 3 desktop environment was officially released in 2011, and in 2020 we are still on version 3.x. Yeah, despite many massive changes over the last (almost) decade, we have been stuck with point releases for GNOME 3. For instance, just last week, GNOME 3.38 was released. Historically, the stable releases all ended in even numbers, with pre-release versions ending in odd. For fans of the DE, such as yours truly, we have simply learned to live with this odd versioning scheme. Well, folks, with the next version of GNOME, the developers have finally decided to move on from version 3.x. You are probably thinking the new version will be 4.0, but you'd be very wrong. Actually, following GNOME3.38 will be GNOME 40. "After nearly 10 years of 3.x releases, the minor version number is getting unwieldy. It is also exceedingly clear that we're not going to bump the major version because of technological changes in the core platform, like we did for GNOME 2 and 3, and then piling on a major UX change on top of that. Radical technological and design changes are too disruptive for maintainers, users, and developers; we have become pretty good at iterating design and technologies, to the point that the current GNOME platform, UI, and UX are fairly different from what was released with GNOME 3.0, while still following the same design tenets," says Emmanuele Bassi, The GNOME Foundation.
Rapid Release? (Score:1)
Also known at CADT (Score:4, Insightful)
This has been called the CADT model for about 17 years now. And I believe that GNOME was even the original source for the term.
Re: (Score:3, Insightful)
JWZ [jwz.org] coined the term, referring to the transition from GNOME 1.4 to 2.x.
Gnome? (Score:1)
Who the hell wants a smelly foot on their desktop?
Re: (Score:2)
Re: (Score:2)
TBH this guy sounds like a bit of a dick. He filed a bug report for an old version, someone re-wrote it from scratch and now he wants THEM to go check if his bug report is still relevant.
On an open source project where people donate their time and energy for free. Where he has no inclination to fix the problem himself.
One of the worst aspects of working on open source software is entitled asshats expecting you to give commercial grade support to their not-so-humble requests.
Re: (Score:1)
You might want to go back and look at who JWZ is. If you're reading this with Firefox, you're using his code. https://en.wikipedia.org/wiki/... [wikipedia.org]
Re: (Score:2)
following GNOME 3.38 will be 40.
Makes sense, I'd take a .40 P320 over a 38 snub nose any day.
If the release is on cadence rather that features (Score:2)
Re: (Score:2)
How about version 2020-04 (for April 2020)? It's partially ISO 8601 compliant, everybody who knows how to read standard/international dates should be able to understand it.
Re: (Score:2)
Re: (Score:3)
Who's call should it be? Can the industry at least standardize on software versions numbering?
"Windows 10 forever" doesn't tell me anything.
Sometime in the year 2055...
Guy 1: What OS is that?
Guy 2: Windows 10.
Guy 3: Wasn't that relased in 2015 or something? Why are you still using an OS that's four decades old?
Note that "macOS 10.15" doesn't tell me much either, but "macOS Big Sur" tells me even less. You're asking people to remember a non-linear list of names based on landmarks of the state you live in? Th
Re: (Score:2)
Windows 10 is not a 'version', it is a product name. There are many versions of Windows 10 (current is 2004). Even if they named it Windows 2015, that would still be the current product, with the same number of versions. 'Windows 2015' tells you exactly as much about how 'outdated' your software is as 'Windows 10' does. In both cases the only way to find out if you are current is to look up what the current actual 'version' is, and compare it to your version.
Re: (Score:2)
Microsoft had to abandon incremental version numbers or they would have been screwed when they reached Windows 94.
Re: (Score:2)
Windows One Series S.
Yeah, I agree. They're bad at naming things.
Re: If the release is on cadence rather that featu (Score:2)
No. Having a version number that references only the date is fine for some projects, but not for others.
Semver version numbers gives much more information than a date based one. You know what the differences are between two versions: patches, features without API breaks, major changes with API breaks.
It also makes it easier for distribs to manage them, often offering a dedicated package for major versions, making it possible to install several major versions and keep all of them up to date.
For projects like
Are they trying to catch up with Chrome and FF? (Score:4, Insightful)
Given that almost 10 years without a new major release hasn't really caused any issues, except for bloat, incredibly horribly enormous bloat.
I really liked Gnome 2, it was loads better than X11/CDE/KDE/OpenWindows... On my desktop Sun.
Then came 3.x. I ditched it for XFCE then Cinnamon about 8 or 9 years ago at home.
Re:Are they trying to catch up with Chrome and FF? (Score:5, Informative)
"hasn't really caused any issues" except for the requirement of systemd, making it near impossible to get it to run on other OSes or distros that don't have systemd. That alone should have been a version bump.
Re: (Score:2)
To be fair that also means that for quite a few OSes you can't avoid systemd even if you don't even run a graphical desktop, much less that specific graphical desktop.
Re: (Score:2)
There's a difference between a distribution's choice of moving to a different init system, and a distribution being forced to due to a major and widely used piece of user facing software depending on that init system.
Now on the flip side you can shim it. Gnome doesn't depend on systemd as much as it depends on functionality that systemd provides and it no longer maintains its kludgy consolekit as a result. The sane approach would be for another maintainer to come in and provide the functionality Gnome needs
Re: (Score:2)
That's a feature not a bug. So anyone sane doesn't have to worry about either one infecting their systems.
Somehow the Molotov-Ribbentrop pact comes to mind...
Re: (Score:2)
Then came 3.x. I ditched it for XFCE then Cinnamon about 8 or 9 years ago at home.
same here.
i'm always fascinated by people writing desktop, desktop apps in javascript and at the end of the day wonder "why is this system requiring more resources?"
Re: (Score:3)
MATE here.
Who do they think they are deceiving? (Score:1)
Re: (Score:2)
Re: (Score:2, Insightful)
People hate Gnome3 so much because it was such a change for the worse from the very popular and cherished Gnome2. It feels like a bunch of haughty morons stole and ruined something that people loved. Like Disney and Star Wars.
Re:Who do they think they are deceiving? (Score:4, Insightful)
Exactly. It catered to everything bad about computing in the last 10-15 years. Lock in to a forced usage model and geared toward devices that treat the owner like a bad actor. Geared toward consumption, not creation. And had the gall to think they could do the user interface better and change something overnight that evolved to what it is over decades. Made bunches of stupid usability decisions that don't line up with real world usage models. And OMG.. the size of icons and wasted space on those systems.
Re: (Score:2)
Lock in to a forced usage model and geared toward devices that treat the owner like a bad actor.
I mean in its defense we have an overwhelming amount of evidence that owners are bad actors. We make stupid decisions against our own best interests.
Re: (Score:2)
Re: (Score:2, Flamebait)
Why do people do this? Don't like it, don't use it. Simple. I'm sure people could point out what they don't like about KDE, Cinnamon, etc. but they don't. They just use Gnome or whatever else and carry on. But no, some AC has to come in and just shit all over someone else's work they do FOR FREE.
Okay, I'll bite. I get that Gnome devs do what they do for free, and I'm very grateful for what Gnome used to be in the GTK2 days. But I'm sure lots of even reasonable, grateful folks like you spend at least a little bit of time complaining about some of the various stupid, vain things done by people who otherwise make valuable contributions.
A commitment to doing away with viable support for mice and non-touchscreen devices, and an apparent desire to see them go away altogether, is stupid and vain - yet tha
Not this debate again... (Score:2)
When Microsoft started doing the whole "Windows [insert-year-of-release-here]" (Ironically the version before '95 being 3.11, like Gnome's 3.x), my non-tech-savvy dad asked me in 1998 when they released Windows '98, "What happened to Windows '96 and '97?"
Why can't we just keep versioning schemes the way they are supposed to be; the way Gnome does it now? Technically speaking it works. Why break it or artificially inflate it like a penis pump?
Re: (Score:2)
If I'm telling you I'm running version X.y of some software, it doesn't tell you how old and potentially out-of-date it is. Even with the confusion you described above, it's easier for regular people and for us (helping them) to use "Software_name version year_of_release".
Re: (Score:3)
If I'm telling you I'm running version X.y of some software, it doesn't tell you how old and potentially out-of-date it is.
Yes it does. It tells you exactly that, that's what traditional version numbering designed to do.
Re: (Score:3)
ok, so I have a software package at realease 5.1. How old is it? Oh, you can't tell from the version number--you would have to lookup the release list for that package? Then you can't tell how old it is.
Re: (Score:1)
If you think that looking up version history for a piece of software is too much work, then you should be re-thinking your field of expertise.
Re: (Score:2)
What does chronological age have to do with anything? What matters is how far out of date it is. If the current version of something is '2020' and your version is '2015', how far out of date is your version? If there were no versions between 2015 and 2020, you are only one version old. If there were 5 versions, you are 6 versions out of date. You would have to look up the release history to determine that.
On the other hand, if the current version is 1.20, and you are at 1.15, you know exactly how out o
Re: (Score:2)
Huh? How could you be 50 versions out of date? You think they count 15, 151, 152, ... 159, 16, ..., 199, 20? Who does that? Just because the minor version is to the right of a dot doesn't mean the minor version is fractions of the major version. It is an integer.
Windows 2015 tells you exactly nothing. Yeah, it COULD be '5 years' behind. Or it COULD be the current version. Or it COULD by anything in between.
Windows 2015 is not a 'version', it is a product brand. Within the Windows 2015 brand there c
Re: (Score:1)
https://en.wikipedia.org/wiki/... [wikipedia.org]
Example: major.minor[.maintenance[.build]] (example: 1.4.3.5249)
Re:Not this debate again... (Score:5, Insightful)
And that still tells you way more than some stupid year-based 'version' tells you.
Major number change - probably a new product altogether, or has compatibility issues with previous versions
Minor number change - feature adds that don't break compatibility
Maintenance change - bug fixes
While it may not show exactly how many 'versions' you are missing, it DOES give you information about the types of changes. A year-based 'version' tells you exactly nothing.
Re: (Score:2)
The whole major.minor.maintenance version used to be true over a decade ago, but Google pretty much killed that with Chrome, then Firefox and others just followed suit. Nobody uses that versioning system anymore because people are used to seeing new major versions every month or so.
Re: (Score:2)
A year-based 'version' tells you exactly nothing.
Actually it tells you what year something came out. This is quite relevant in the modern rolling release model where the major / minor regime ceases to make any kind of developmental sense anymore.
Re: (Score:2)
What are you talking about? I gave the exact same information based on year: If the current version of something is '2020' and your version is '2015',
Re: (Score:3)
Software using version numbers... we are on release 2.0.11... .0 releases, tell me when you have 2.1 out. ... ... wtf?
customer: I don't use
Us:
Us: We are changing our naming scheme to 20.04.01.11 to reflect the date
Customer: That works. We will try it out now.
Us:
Yep, it actually happened. A customer insisting they won't use a .0 release (despite a patch number) triggered us to use a timestamp based version format.
Re: (Score:2)
I see no reason to bend the way you do versioning for the user. User education should happen instead.
Re: (Score:2)
Re: (Score:2)
He said customer not just user. When you have customers you do what makes money
You just described software companies that put more money into marketing than into development.
Re: (Score:3)
Wait, doing what the CUSTOMER wants is 'marketing'? Since when. Getting the customer to do what YOU want is marketing, and that is the approach YOU are suggesting.
Re: (Score:2)
No, what I'm saying is that changing the way you version your software code isn't something customers should decide. If it is, you might as well ask them what IDE they want you to use, or what language for that matter.
Re: (Score:2)
I apologize. I originally read the OP as they changed the versioning to match what the customer wanted. Now I see they did exactly the opposite, they changed to deceive the customer. That is obviously wrong.
Open source shenanigans as usual (Score:2)
Plotting and scheming their versions numbers.
Um ... (Score:2)
Actually, following GNOME3.38 will be GNOME 40. "After nearly 10 years of 3.x releases, the minor version number is getting unwieldy.
Um, 38 is unwieldy? I know it's past the number of fingers and toes, but seriously? :-)
I'm hoping this does not explain a lot about GNOME.
Re: (Score:2)
Actually, following GNOME3.38 will be GNOME 40. "After nearly 10 years of 3.x releases, the minor version number is getting unwieldy.
Um, 38 is unwieldy? I know it's past the number of fingers and toes, but seriously?
I'm hoping this does not explain a lot about GNOME. :-)
What? Real Professionals count in binary! which is about 1.04 million, as long as you have all your fingers and toes, and (most men) can get up to 2^21.
Re: (Score:2)
I can get to 1024 on my fingers alone and 1,048,576 if I use my toes too.
Idiots (Score:5, Insightful)
XX.YY.ZZ
XX: changes that break backward compatibility.
YY: changes that add functionality without losing backward compatibility.
ZZ: bug fixes.
What is so bloody hard about that? I know it's too complicated for the Python devs to follow, but for most normal people it's simple and informative.
Re: (Score:2)
This times 1000. It is quite frustrating that people time and time again believe they have invented a better, more "rounder" wheel. It is these types of decisions that tell me...nay scream to me there is no true professional running the Gnome project.
Re: (Score:3)
Backward compatibility in a desktop environment? Are you drunk? When your product is a library with a programming interface you should never break backward compatibility. When you are writing something that isn't exactly a library, the concept of backward compatibility doesn't make any sense. So basically your dumb scheme doesn't make any sense, ever.
Re: (Score:2)
Backward compatibility in a desktop environment? Are you drunk? When your product is a library with a programming interface you should never break backward compatibility. When you are writing something that isn't exactly a library, the concept of backward compatibility doesn't make any sense. So basically your dumb scheme doesn't make any sense, ever.
You are the sort of idiot I was talking about. Too dumb to know you're dumb.
Re: (Score:2)
Ran out of arguments?
Re: (Score:2)
This is really stupid. You do realize there is a hell of a lot more to 'compatibility' than just 'programming interfaces', don't you? Like, do the users have to change what they have been doing? Is new training required? Is any work done by the user still valid? Does the program still interact with any other programs in the same way (can a file created by your program still be used with some other program)? You know, minor shit like that. Or are you one of those 'screw the user, developers are all th
Re: (Score:3)
News flash.. gnome is built on top of a LOT of libraries.
Re: (Score:2)
So what? Is GTK going to adopt the Gnome versioning scheme? That would be some news and completely retarded.
Re: (Score:1)
> When your product is a library with a programming interface you should never break backward compatibility.
Turn out people do break backwards compatibility with minor changes. Some idiots push their preferences into defaults without considering that their defaults break code execution on slightest deviation from specific data input format.
Re: (Score:2)
That's one reason why Windows' UI is so shit, they won't break backwards compatibility and are stuck with things like the stupid way the Start Menu works.
Re: (Score:2)
So if something never break backward compatibility, it's condemned to never change its major version number? How is that not idiotic?
Re: Idiots (Score:2)
Re: (Score:2)
What is so bloody hard about that?
Nothing providing you don't do rolling releases.
Cinnomin is Everything Gnome Should Have Been (Score:2)
serious (Score:1)
When (Score:2)
WHen came it catch up to 2010 Gnome 2 and not be a useless utter piece of crap right up there with Nautilus which makes Window 8 look usable with Gnome's cell phone interface
then, nvidia-numbering? (Score:2)
Also Known in CADT (Score:1)
Re: Also Known in CADT (Score:1)