Plastic 2.0 version control is out!
pablo
I'm very proud to announce Plastic 2.0. Check what's new at www.plasticscm.com
We've been working very hard during the last year to release this new
version. Here are some of the new features:
- Distributed system: now Plastic can work as a distributed SCM
system. You can replicate changes back and forth and merge them when
there are conflicts. This functionality opens up a number of different
scenarios: from fully distributed development (each developer running
his own plastic server) to multi-site support, whether it is based on
mastership or not.
- Multi-server support: Plastic server's can be configured to work in
workspace mode, repository server mode or both. The default is both,
but sysadmins can choose to set up a more complex network of plastic
servers for better scalability and performance.
- New and improved GUI: we've totally redesigned the user interface.
Now it runs entirely on both Windows and Linux, and it has an improved
look and feel. We feature a multi-view interface which is something
our users were demanding for the last months.
- Code review helper tool: to inspect changes made on changesets,
branches and labels. It really helps to review changes made by the
other developers.
- Query system: almost all the views in the GUI can be customized
using a query language (also available from the command line) which is
very similar to SQL. It is very flexible and can be used to create
customized reports too.
- MySql backend: plastic stores all data and metadata in standard sql
backends. So far we supported SQL Server and Firebird, and MySql has
been our latest addition. It introduces enhanced performance and
scalability.
- Statistics module: the new GUI comes with a statistics module which
lets you explore who made changes to which elements.
Hope you enjoy the new version of Plastic SCM.
It is FREE for educational purposes and open source projects.
Pablo
marlow...@googlemail.com
> Hi all,
>
> I'm very proud to announce Plastic 2.0. Check what's new atwww.plasticscm.com
> It is FREE for educational purposes and open source projects.
This is a closed source commercial product. By 'free' the OP means
currently released free of charge.
The question that springs to mind (apart from why is it not open
source) is why another SCM? What about Mercurial? Monotone? Git? What
advantages does plastic have over the open source SCMs?
-Andrew Marlow
Bruce Stephens
[...]
> The question that springs to mind (apart from why is it not open
> source) is why another SCM? What about Mercurial? Monotone? Git?
> What advantages does plastic have over the open source SCMs?
Don't know about Plastic SCM specifically, but generally you might
expect a more polished product (with GUIs and things that all work
properly), support that you pay for.
Free SCMs don't seem to offer particularly good issue-tracking for
whatever reason, so perhaps that's something Plastic offers. A quick
read of the web site suggests that Plastic uses a different model of
branching (maybe closer to AccuRev's streams?). No idea whether
that's actually important, but it's a difference.
Marc Girod
> This is a closed source commercial product. By 'free' the OP means
> currently released free of charge.
Open *source*, such as in subversion, for instance, doesn't
buy you much more *practical* freedom.
It is also very much 'free of charge' only.
Consider the effort there is to publish changes of yours
to a system which only supports *in practice* a /main/LATEST
model, and therefore restricts write access to 'committers',
with the effort there is to --say-- patent them.
> The question that springs to mind (apart from why is it not open
> source) is why another SCM? What about Mercurial? Monotone? Git? What
> advantages does plastic have over the open source SCMs?
I won't go into the case of Plastic.
But non withstanding the good bits of the 3 tools you name
(and I like particularly the implicit branching of mercurial),
there is a *humongous* space for new ideas in SCM.
Those tools in particular are nothing else than source
control tools, which is utterly inadequate especially for
the needs of Open Source.
The lack of proper SCM (Software Configuration Management
btw) is properly the reason of the failure of Open Source (and
of Free Software) in the last 10 years.
The reason why the commercial world (including RedHat and
Collabnet) has been able, with 'free of charge' software, to
take back the lead (Google...).
http://www.cmcrossroads.com/cgi-bin/cmwiki/view/CM/SubVersionFree
SCM has gone backwards in the last 10 years.
It has never been so low.
Users do not expect anything from SCM tools.
They pick tools mainly on the basis of their 0 price tag.
Marc
Marc Girod
wrote:
> Don't know about Plastic SCM specifically, but generally you might
> expect a more polished product (with GUIs and things that all work
> properly), support that you pay for.
GUIs are, at the current level of the technology, an SCM killer.
They introduce discontinuities, dependency bottlenecks (they
blur in presentation concerns).
They make it harder to trace causality chains.
They make it harder to communicate, by giving users a false
sense that the communication has already happened, by
trapping them into a feeling of obviousness.
Support is usually something you want to escape.
It introduces an additional layer of opacity, with diverging
interests. The fact that you pay is no value in itself.
Quantitative factors typically hide qualitative ones.
> Free SCMs don't seem to offer particularly good issue-tracking for
> whatever reason, so perhaps that's something Plastic offers.
I have not seen any good commercial issue-tracking
product either. Again, discontinuities, noise blurring
the signal, with artificial ticket boundaries, idiotic
concerns of meaningless work-flows: tickets closed
while the cases are not really solved. Pressions of
all kind, ignorance, carelessness...
> A quick read of the web site suggests that Plastic uses
> a different model of branching (maybe closer to AccuRev's
> streams?). No idea whether that's actually important, but
> it's a difference.
From the discussions with Pablo so far, nothing
unique or really novel. Consistent at least.
http://www.cmcrossroads.com/cgi-bin/cmwiki/view/CM/TokenPassingTools
http://www.cmcrossroads.com/cgi-bin/cmwiki/view/CM/IgnoranceIndifference
Marc