views:

287

answers:

5

I'm looking for "schemes" for application version or milestone naming.

I'd like to find a naming scheme that suits my current project but I think we should gather here the more interesting schemes to share knowledge and ideas on the subject.

You also should provide the number of possible name with the proposed scheme or an estimate.

For example :

The development codename of a release takes the form "Adjective Animal". So for example: Warty Warthog (Ubuntu 4.10), Hoary Hedgehog (Ubuntu 5.04), Breezy Badger (Ubuntu 5.10), are the first three releases of Ubuntu. In general, people refer to the release using the adjective, like "warty" or "breezy".

That allow a finite but very big possible name count.

  • Some hardware and software releases are named like real world big cities - hundreds of possible names;
  • Some hardware and software releases are named like (Greek-Roman-whatever) mythology gods or king/warlord/emperor - tenths of possible names my mythology;
  • etc.

For my own project I'm thinking about using a naming scheme that would be to name public releases like popular software principles/philosophies/rules/pattern like DRY, KISS, YAGNI, Singleton, Factory, Visitor, etc. But it feel maybe a little too much...computer-geek. :)

+1  A: 

We used to use all the different names from hell and its mythology. Our production server was my favorite: Horus.

Tom Anderson
+1  A: 

How about using collective noun pairs (apostrophes left out deliberately):

aardvarks armoury

blacksmiths anvil

crow murder

tanks clank

snipe wisp

Mitch Wheat
+4  A: 

The problem with programming terms as project names is that they're confusing for most users. They'd make great internal codenames.

I've heard major rivers used as project names - Nile, Brisbane, etc.

check out http://en.wikipedia.org/wiki/List_of_rivers

Releasing in alphabetical order helps users remember which version is more recent.

akatakritos
Having internal code names and separate public release names is interesting, thanks for the suggestion! Now that I think about it, computer software development popular principles and patterns could be confusing in the devs team chats...
Klaim
+1  A: 

frankly i don't get Ubuntu's release naming scheme - too many names thrown at us too often. Windows can get way with XP and Vista every now and then (every 3-4 years?) but Ubuntu is too aggressive with hodgepodge of names every 6 mos.

to answer your question - what's wrong with good old 8.10 and 9.04? KISS indeed

webwesen
The version number scheme is not a problem and is already set. The problem is more to give "human-friendly" names to my software's public releases (one for several versions numbers), for easier user feedback (the name is displayed in big chars inside the application) and subtle marketing purposes...
Klaim
e.g. like Apple's OS scheme - 10.4 = Tiger, 10.5 = Leopard etc. You can use either name and most people are familiar with both
Phil Nash
after a while these names do not mean much - was "Mouse" before the "Elephant"? or the other way? whereas 10.4 and 10.5 leave no ambiguity
webwesen
+1  A: 

I personally find naming release versions of a software product after mythological creatures, furry animals and geographic landmarks all just a bit new age and pretentious. Fine for a beta or internal working codename maybe, but there's no substitute for proper version numbers when releasing to your end users.

Just my 2c
Kev

Kev