• bearophile <bearophileHUGS lycos.com> May 15 2009
• Daniel Keep <daniel.keep.lists gmail.com> May 16 2009
• Jarrett Billingsley <jarrett.billingsley gmail.com> May 16 2009

bearophile <bearophileHUGS lycos.com> writes:

Leandro Lucarella:

I think this is another problem with D, version naming is really confusing and
lame. You can't know anything from a D version number.<


Yes, improving such small things is positive.
So I suggest to start using a "language.version.releaseStatus" numbering scheme
for D2 (and maybe for D2 too).

So the current D2 becomes:
2.0.30alpha

and the current D1 becomes:
1.0.45

Once D2 gets out of alpha it may become:
2.1.0

Bye,
bearophile

Daniel Keep <daniel.keep.lists gmail.com> writes:

bearophile wrote:
 Leandro Lucarella:
 
 I think this is another problem with D, version naming is really confusing and
lame. You can't know anything from a D version number.<


 Yes, improving such small things is positive.
 So I suggest to start using a "language.version.releaseStatus" numbering
scheme for D2 (and maybe for D2 too).
 
 So the current D2 becomes:
 2.0.30alpha
 
 and the current D1 becomes:
 1.0.45
 
 Once D2 gets out of alpha it may become:
 2.1.0
 
 Bye,
 bearophile


I don't think this is sufficient.  What we really need is to treat each
part of the version as a complex number.

In this way, non-stable releases can have imaginary components to
distinguish them from release versions.

So the current series of D 2.x compilers would become D 2i.x.  Release
candidates would be D 2i.xi.  And of course, a stable release which has
unreleased modifications could be D 1.x+yi

So much more intuitive than the current system plus a big, red label
reading "WARNING: not stable, do not use."

  -- Daniel

Jarrett Billingsley <jarrett.billingsley gmail.com> writes:

On Sat, May 16, 2009 at 11:07 PM, Daniel Keep
<daniel.keep.lists gmail.com> wrote:
 I don't think this is sufficient. =A0What we really need is to treat each
 part of the version as a complex number.

 In this way, non-stable releases can have imaginary components to
 distinguish them from release versions.

 So the current series of D 2.x compilers would become D 2i.x. =A0Release
 candidates would be D 2i.xi. =A0And of course, a stable release which has
 unreleased modifications could be D 1.x+yi


Why restrict yourself to a two-dimensional complex plane when we have
quaternions and octonions?