Board index » cppbuilder » Re: A recomendation for reasonable version control system.

Re: A recomendation for reasonable version control system.


2005-02-26 06:03:36 AM
cppbuilder75
Boian Mitov wrote:
Quote

What about Canada and US? So far that is all we do.
I think they're the same 'zone', so no problem there.
Don't take my word for it, though :-)
--
Kristofer
 
 

Re:Re: A recomendation for reasonable version control system.

Boian Mitov wrote:
Quote
You are way off. I am talking about file share between library
location and install location as example. You can't avoid that, can
you? Please think before saying something. File share across projects
is a must! And there is no acceptable alternative.
Sorry Boian, my wording was not very friendly.
I am frankly not familiar with the scenario you mention, but I can
imagine there are good reasons for it.
Your earlier post (quoting: "vast amount of files, well within millions,
most of them binary, and we change and rearrange the directories and sub
directories on daily basis") made the whole development process seem
extremely fluid, I guess that's what sparked my comment. I've never
worked on a project with such characteristics. It would be interesting
to hear the drivers behind this "dynamo" of ever-mutating files and
directories.
With regards to file sharing, for sure files need to be shared among
projects, however in my environment these library files are more
strictly controlled in discrete and formally tagged releases; that is,
you either use "version 1" of Library_File_A, or "version 2", but never
something-in-between, or whatever-the-latest-checkin-was. So what we
usually end up with is a "configured library" tree of read-only
checkouts on our network. People can only change these files going
through a special qualification cycle and approval with the
"Librarian" - after which, the new release is deployed to the network
library tree.
As you may have surmised from my other posts, I am also 'on the prowl'
for a new version control system (currently half-heartedly living with
CVS). And one of the main things I want to figure out is the
*limitations* of the various systems out there, so that I can be
somewhat confident that I won't get stuck with a hidden 'gotcha' a few
months down the road.
--
Kristofer
 

Re:Re: A recomendation for reasonable version control system.

Boian Mitov wrote:
Quote
Not until we get a working Java. I have worked with SmartCVS in the
past. What a nightmare. Worse than CBX. Not to mention that it is not
compatible with CBX, and as soon as you install the one the other gets
broken :-( . Java!
LOL, as long as you don't use *any* other Java tools then you should be
fine <g>.
Hmmm, I have a copy of MagicDraw lying around here... <sigh>well, good
riddance, I'm not using it anymore.
But apart from the compatability thing, how bad was SmartCVS?
--
Kristofer
 

{smallsort}

Re:Re: A recomendation for reasonable version control system.

Hi Kristofer,
It broke before even I got well enough with it :-( . As it was more
important to have CBX at the time it was s{*word*99}ped. It did not offer the
file sharing ( a CVS fault I guess ), so there was no reason to keep it
alive. It was not a very friendly UI however, with a lot of Java
flickering and menu misbehaviors. Those may have been fixed since then.
Cheers,
Boian
Kristofer Skaug wrote:
Quote
LOL, as long as you don't use *any* other Java tools then you should be
fine <g>.
Hmmm, I have a copy of MagicDraw lying around here... <sigh>well, good
riddance, I'm not using it anymore.
But apart from the compatability thing, how bad was SmartCVS?

 

Re:Re: A recomendation for reasonable version control system.

We seems to be on the same boat here ;-) . I have been half-heartedly
living with VSS for a while ;-) . I am testing them all one by one. A
very tedious experience :-( . Long live the VMWare ;-) . I can even test
Java stuff without messing up my system ;-).
Cheers,
Boian
Kristofer Skaug wrote:
Quote
As you may have surmised from my other posts, I am also 'on the prowl'
for a new version control system (currently half-heartedly living with
CVS). And one of the main things I want to figure out is the
*limitations* of the various systems out there, so that I can be
somewhat confident that I won't get stuck with a hidden 'gotcha' a few
months down the road.

 

Re:Re: A recomendation for reasonable version control system.

Hi Kristofer,
Thank you! I got it to work yesterday. It has some little hiccups
when synching. Also have not been able to share a file across projects
or dirs. I have sent then e-mail. No respond yet. :-( . I will keep you
informed what I have discovered. I am also messing with StarTeam in a
Virtual Machine. Didn't dare to install it on the primary one, as I
suspect that it may have some Java in it ;-) . Until I am sure I can
install at least the client without any Java it will stay on Virtual
Machine.
Cheers,
Boian
Kristofer Skaug wrote:
Quote
You can! (I just did, last week); just install the server locally and
the client.
Works here... can't remember where the 'catch' was, problably
time-limited
before it requires registration or something.
FWIW, I think the TC user interface is refreshingly clean and
'un-geeky',
it hits the sweet spot of balance between 'too much eye candy' and 'too
freakin' many techie gadgets in default view'.
I hadn't realised it didn't do 'deltas', that'd be a major drawback.
We still haven't decided on our new version control tool.

 

Re:Re: A recomendation for reasonable version control system.

Wasn't the BitKeeper the p2p one? I relay am a bit scared to run a
p2p version control system.
Cheers,
Boian
Kristofer Skaug wrote:
Quote
Oscar Fuentes wrote:

>People talks about BitKeeper as a truly excellent VCS:


They do indeed. But I was disappointed; it *looks* too much like
a disjointed collection of a unix-head's private hacked-up stuff.

 

Re:Re: A recomendation for reasonable version control system.

Boian Mitov wrote:
Quote
Didn't dare to install it on the primary one,
as I suspect that it may have some Java in it ;-) .
LOL, you remind me of the Monty Python 'Spam' sketch:
bau2.uibk.ac.at/sg/python/Scripts/TheSpamSketch
<quoting>
Wife: Have you got anything without spam?
Waitress: Well, there's spam egg sausage and spam, that's not
got much spam in it.
Wife: I don't want ANY spam!
</quoting>
--
Kristofer
 

Re:Re: A recomendation for reasonable version control system.

Boian Mitov wrote:
Quote
It was not a very friendly UI however, with a lot of
Java flickering and menu misbehaviors. Those may have been fixed
Drat. I had been sort of hoping that SmartSVN might turn out to be a
usable UI (hadn't had time to look further than the front page, noting
it was still in beta). I *like* almost everything SVN is advertising so
far, except the lack of a decent UI. Tortoise isn't half enough, it
would have to be a CvsGUI (for SVN) but preferably more.
--
Kristofer
 

Re:Re: A recomendation for reasonable version control system.

I got the same impression from SVN. No GUIs :-( . And I am still not
sure it supports file share. Other than that sounds great.
Cheers,
Boian
Kristofer Skaug wrote:
Quote
Drat. I had been sort of hoping that SmartSVN might turn out to be a
usable UI (hadn't had time to look further than the front page, noting
it was still in beta). I *like* almost everything SVN is advertising so
far, except the lack of a decent UI. Tortoise isn't half enough, it
would have to be a CvsGUI (for SVN) but preferably more.

 

Re:Re: A recomendation for reasonable version control system.

"Boian Mitov" < XXXX@XXXXX.COM >wrote in message
Quote
Didn't dare to install it on the primary one, as I suspect that it may
have some Java in it ;-) . Until I am sure I can install at least the
client without any Java it will stay on Virtual Machine.
Cheers,
Boian


So, to avoid installing the Java Virtual Machine, you install 80x86 Virtual
Machine... :(
If I start a rumour that MS Virtual PC is actually written in Java, then
Boian will be really stuck...
- Roddy
 

Re:Re: A recomendation for reasonable version control system.

Quote
FWIW, many of our larger (thousands of users) StarTeam customers moved
to StarTeam because it was more performant than any other version
control system they evaluated.
<microrant>
But obviously, less Englishant..?
I assume you were trying to say it either "performed better" or "ran
faster". Both these phrases have the advantage that they're easily
understood, shorter, and use words that actually exist in dictionaries.
</microrant>
- Roddy
 

Re:Re: A recomendation for reasonable version control system.

Roddy Pratt wrote:
Quote
But obviously, less Englishant..?
Sorry you don't like "performant".
Quote
and use words that actually exist in dictionaries.
What dictionary did you look in?
I have a very large vocabulary. Sometimes it manifests itself in my
posts when I'm writing quickly.
--
John Kaster blogs.borland.com/johnk
Features and bugs: qc.borland.com
Get source: cc.borland.com
If it's not here, it's not happening: ec.borland.com
 

Re:Re: A recomendation for reasonable version control system.

Hi Ruddy,
I am actually using VMWare ;-) . And I am only testing to see if the
sucker I downloaded is going to install Java VM or not ;-). Then again,
if the VMWare becomes written in Java, it indeed will be a different
Java, the one worth being installed by me ;-). Unfortunately we don't
have such Java yet, now do we? ;-) So far VMWare have not crashed my
system. I wish I could say the same for Java. I had crashes even from
web Applets in the past.
Cheers,
Boian
Roddy Pratt wrote:
Quote
So, to avoid installing the Java Virtual Machine, you install 80x86 Virtual
Machine... :(

If I start a rumour that MS Virtual PC is actually written in Java, then
Boian will be really stuck...

- Roddy


 

Re:Re: A recomendation for reasonable version control system.

"John Kaster (Borland)" < XXXX@XXXXX.COM >wrote in message
Quote
What dictionary did you look in?
Concise Oxford Dictionary, 1976 edition (the most recent I have to hand!) No
sign of 'performant' at all.
But, dictionary.com offers 'performant' as a noun meaning performer (as in
informer/informant), but no hint of it as an adjective.
However, google has 100K+ hits for 'performant', so it's obviously in
regular use somewhere!
Quote
I have a very large vocabulary. Sometimes it manifests itself in my
posts when I'm writing quickly.
No problem there - I have learnt a new word today, even if it's not one I'd
naturally use myself. And dictionaries change to reflect current usage, not
vice-versa.
How about 'performanter' (as in "Delphi is performanter than BCB"?)...
Regards,
Roddy