[License-review] Fwd: [Non-DoD Source] Resolution on NOSA 2.0

Rick Moen rick at linuxmafia.com
Sat Jun 23 07:41:48 UTC 2018


Quoting Karan, Cem F CIV USARMY RDECOM ARL (US) (cem.f.karan.civ at mail.mil):

> And I just thought of another point;  OSI is a fairly long-lived
> organization, and is definitely older than Git*.  if Git* goes under,
> what happens to the issues, etc. on Git* servers? The git repositories
> would be reasonably safe due to everyone pulling from them
> continuously, but there is a lot of other metadata that could be lost.
> It would be better to have everything under git (or other DVCS) so
> that there are enough copies around the net that if one person or
> organization dies/goes down/whatever, then the rest of the network can
> recover and rebuild.  

Your assessment seems like a good one.

Having seen types of Internet service come and go, and likewise 
particular hosting instances come and go, it's been apparent that only
data in fairly simple commodity formats tend to persist online more
than a few years:  When technology changes for more-sophisticated 
data formatting, as it always does, the changeover is lossy.

That is in _no_ way intended as an argument against adopting an issue
tracker.  Mailing lists have many virtues, including support for very
long collective memory (e.g., I trivially extended Silicon Valley
Linux User Group's Mailman mailing list archive an extra year, not long
ago, back to September 1997, by importing the earlier Majordomo
incarnation's mbox -- so 21 years of continuous records and counting),
but they're pretty terrible at the things for which issue trackers
excel.

Fossil looks intriguing at initial glance, hitting IMO a good middle
ground for features, a small/modest/fast engine that might survive
security auditing, and having no horrible misfeaures I can spot --
although writing a new, one-off DVCS in a world with git, mercurial,
monotone, darcs, and bzr seems quixotic.  (Author says existing DVCSes
just didn't meet his needs.)

I had some difficulty taking a closer look at Veracity because almost
all of the information links on its Web site are 404, probably why
Wikipedians comment 'Web site appears unmaintained'
(https://en.wikipedia.org/wiki/Comparison_of_version_control_software).
Ceteris paribus, that's a bad sign.  Also, there have been no releases
for five years, and I see no development code at all other than a
non-functional-for-me link to nightly builds.  Also, the Veracity-users
mailing list that once existed 'is no longer active' (per the project
pages) and the links to past archives at Gmane.org are dead because
Gmane discontinued its Web interface in July 2016.  Veracity looks to me
like a SourceGear, LLC product that firm produced 5-7 years ago behind
closed doors without community involvement, and more than likely is no
longer maintained at all.  (There are code tarballs under Apache Licence
2.0, on the other hand.)

(Apologies for not being immediately able to suggest others.  I'll see
if I can find some.)

-- 
Cheers,                                           "A recursive .sig
Rick Moen                                         Can impart wisdom and truth.
rick at linuxmafia.com                               Call proc signature()"
McQ! (4x80)                                       -- WalkingTheWalk on Slashdot



More information about the License-review mailing list