Quotes of the week

[Posted January 18, 2012 by corbet]

One always got the feeling that somebody was steering GNOME, but it wasn’t clear who. When it started, I thought it was Miguel and Nat, then Novell, then Redhat. Now it has that floaty, determined meandering that the best mass open source projects have. From a distance, everyone seems to be constantly bickering and regretting the next steps; but the steps get made, and slowly everyone adapts to them. GNOME feels like a nation now.

-- Danny O'Brien

But these days anything is possible with version-numbers really, except for going backwards. Which is precisely what we are avoiding here.

Just look at Mozilla Firefox (moving from 4 to 9 at the same pace as they went from 0.7 to 1.0) or Google chrome (what version-number are they using anyway?), or the linux-kernel, going from 2.6.0 to 2.6.39 with entire subsystems being rewritten from scratch, and then moving from 2.6.39 to 3.0 without any radical change whatsoever.

Really, moving from 4.8 to 4010 is not really that big a deal, if it serves the right purpose.

-- Stephan Arts for the Xfce project

Quotes of the week

Posted Jan 19, 2012 14:03 UTC (Thu) by nix (subscriber, #2304) [Link] (11 responses)

A package manager that has a problem with 4.10 being a later version than 4.9 has no reason to exist. (I also suspect that no such package managers exist, since this transition has been done numerous times by all sorts of critical packages in the past, with no problems reported.)

Quotes of the week

Posted Jan 19, 2012 14:49 UTC (Thu) by proski (subscriber, #104) [Link]

Perhaps Xfce should use version 5.0 and show an example to KDE and GNOME how a desktop environment can increase its major version number without screwing its users in process of doing so.

Quotes of the week

Posted Jan 19, 2012 16:04 UTC (Thu) by cesarb (subscriber, #6266) [Link] (1 responses)

Follow the link at the very end, the one where it says "Update-2".

(The 4.10 < 4.9 part was the first hint I had that something was not right with that blog post.)

Quotes of the week

Posted Jan 26, 2012 18:47 UTC (Thu) by nix (subscriber, #2304) [Link]

Well, if someone posts an April Fool's joke in January they deserve what they get. It was too plausible to be effective satire, alas.

Quotes of the week

Posted Jan 19, 2012 22:11 UTC (Thu) by mathstuf (subscriber, #69389) [Link] (2 responses)

Doesn't perl/CPAN do float-compares for version numbers? Maybe that's no longer the case.

Well, it USED to be the case...

Posted Jan 20, 2012 5:41 UTC (Fri) by khim (subscriber, #9252) [Link] (1 responses)

But recently¹) they changed that - they use version tuples now: a literal such as v5.6.0, for example, will be interpreted as a Unicode string made up of three characters: character 5, character 6 and character 0.

────────────
¹) if you use geological epochs as primary time unit that is: that happened over ten years ago - that's why after perl 5.005 they had the audacity to release perl v5.6.0 ...

Well, it USED to be the case...

Posted Jan 20, 2012 16:33 UTC (Fri) by mathstuf (subscriber, #69389) [Link]

Ah. I just remember running into a discussion on different versioning schemes when I started packaging and it was mentioned. (I don't do much perl stuff if you can't tell ;) ).

Quotes of the week

Posted Jan 19, 2012 23:38 UTC (Thu) by AndreE (guest, #60148) [Link] (1 responses)

Note he says package maintainers, not managers.

It is still an absolutely ludicrous situation though. Did people really get confused between kernel 2.6.9 and 2.6.39?

Quotes of the week

Posted Jan 20, 2012 4:53 UTC (Fri) by elanthis (guest, #6227) [Link]

I still recall rather vividly an argument some years back over the Evolution mail client. A user on the mailing lists -- and hence almost by definition a more technical user than the average lay computer user -- was throwing such an immense stink over how "stupid" it was for client to transition from x.9 to x.10 because _obviously_ everybody knows that x.10 is equivalent to x.1 and that only incredibly massive morons could ever possibly think that x.10 comes after x.9.

Of course this is one of several reasons why larger successful products use marketing numbers and versions rather than internal ABI versions. The three-component non-decimal numbers make a lot of sense for ABIs. Big round non-fractional non-denominated non-structured version numbers make sense for marketing and releases of large collections of components (such as an OS).

Quotes of the week

Posted Jan 20, 2012 18:02 UTC (Fri) by Cyberax (✭ supporter ✭, #52523) [Link] (2 responses)

Come on. There's no kill like overkill!

My first software will have v120391023 for the _initial_ release. The next version will obviously be v120391023^2. Maintenance releases will be indicated by the number of factorial signs after the version.

So one will get: v120391023^2!, then v120391023^2!! and so on.

Try beat that, Chrome!

Quotes of the week

Posted Jan 23, 2012 10:29 UTC (Mon) by mpr22 (subscriber, #60784) [Link]

Version g₁, g₂, etc.

Quotes of the week

Posted Jan 28, 2012 3:32 UTC (Sat) by nlucas (guest, #33793) [Link]

v.----...---...-- (v1.23)
v.----....--....- (v1.34)

Now a package manager that can cope with that! ;-)

Version numbers

Posted Jan 28, 2012 4:12 UTC (Sat) by douglasbagnall (subscriber, #62736) [Link]

When I was a novice Gnome 2.4 user, I spent months ignoring irritating headlines about old versions like 2.10 and 2.12. Why didn't these people just upgrade? Wasn't their distro as futuristic as Debian?

In those days I was merely numerate. I know better now, but still I would vote for 5.0, thence 5.02 (or, shock, 5.01). What's the harm in being explicable to whole world?