Fossil. [Was: Intent to work on speedups to `clock format` and `clock scan` (#4)]

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Fossil. [Was: Intent to work on speedups to `clock format` and `clock scan` (#4)]

Jan Nijtmans-2
2017-06-02 0:37 GMT+02:00 Dipl. Ing. Sergey G. Brester:
> But even if we speak about...
> Another thing, what always hold me back (resp. disturb by the work and cost
> time) is ... FOSSIL.

Well, your complicated 3-way merge already indicated to me that you're
not familiar with the Fossil way of working. It's different from GIT's, which
- indeed - makes cooperation between the two not trivial (it can be done,
but I wouldn't use import/export for that).

So, you learned to play the piano, but now you are playing the guitar.
Both have strings, both result in beautiful music. It's not easy to play
guitar music on the piano or reverse, unless you are familiar with
both models.

Well, beleave me, playing Fossil is MUCH simpler than playing GIT.
But - indeed - it's different.

Regards,
       Jan Nijtmans

B.T.W.: I play both the guitar and the piano.   ;-)

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Tcl-Core mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/tcl-core
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fossil. [Was: Intent to work on speedups to `clock format` and `clock scan` (#4)]

Alexandre Ferrieux
On Fri, Jun 2, 2017 at 12:12 PM, Jan Nijtmans <[hidden email]> wrote:
>
> So, you learned to play the piano, but now you are playing the guitar.
> Both have strings, both result in beautiful music. It's not easy to play
> guitar music on the piano or reverse, unless you are familiar with
> both models.
>
> Well, beleave me, playing Fossil is MUCH simpler than playing GIT.
> But - indeed - it's different.

> B.T.W.: I play both the guitar and the piano.   ;-)

+1

BTW: I reluctantly learnt the bagpipe (CVS), disliked it, then somehow
came to an agreement with piccolo (SVN). When the wave of DVCS hit us
(Tcl) and me (at work) I was exposed to piano (git) and guitar
(fossil) simultaneously, and forced to use both in these parallel
contexts. Bottom line: the guitar rocks . This may sound like personal
taste and details of ergonomics, but  the analysis is generic and
crystalline:  a cleaner staging model directly translates into "lower
mental load": if the DVCS is not your main focus in life, with Fossil
you can afford to allocate it a very small space in your brain, and
keep working efficiently. With Git you need to "keep it in cache" and
a bigger cache at that. Larger than what I have at least.

-Alex

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Tcl-Core mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/tcl-core
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fossil. [Was: Intent to work on speedups to `clock format` and `clock scan` (#4)]

Dipl. Ing. Sergey G. Brester
In reply to this post by Jan Nijtmans-2
These were not really the arguments to all things from my statement, or?

My "complicated" 3-way merge was at all not complicate. What makes you think that?
It was just made (even to save my own time), because I had previously developed in trunk, and back-ported to 8.6 from trunk (with resolving conflicts and changing of all incompatible, e.g. "int refCount" to "size_t refCount", etc.), and then as it was ready in 8.6 merged it again in trunk (then without conflicts anymore, out-of-the-box).

I known what it in fossil is different from GIT's. Believe me...
I spoke about another things there... If you don't understand (or rather don't want to understand) this, it's good to me. No problem. But don't try to explain it all as "You can just not use fossil"...
Because I'm familiar with both. Just the reasons that were mentioned are other as yours...

As an example (mentioned in my previous mail):

I'm on develop-state in the customer-repo. This is proprietary (not public). This is an complex project in application server and there tcl, tcl-threads. etc. are just a sub-modules (in sense of git).

  • For example, I found some bug (e. g. SEGFAULT), I don't know where exactly.
  • By trying to find it I fixed it directly in modified tcl-code and then notice that the core since 8.5 would be affected (but very hard to reproduce there).
  • I'm currently in 8.6-mod-X in foreign origin.
  • I want to rebase my (extensive) changes to branch based on core-8-5-branch.
  • Then merge "sebres-8-5-cr105" into core-8-6-branch / trunk.
  • Then reintegrate it into several development stands of some customers, to automatically deploy and start the test-cases (I'm lazy and have no time at all to do it manually).
  • Then I found a bug now in my bug-fix.
  • I'll repeat whole process (7x, so cr105-v.1 up to cr105-v.7).
  • I have noticed, that the 5th solution was the best one.
  • Then I'm ready and want to push reintegrated changes back to origin (without make it public, because proprietary).
  • Then I'll reintegrate to origin of tcl.core, I notice the broken test-case there. I repair it and repeat this again.
  • Then I'll reintegrate to origin of tcl.core, to make my changes finally public (in the meantime "cr105-v.8", without proprietary code pieces, etc.).
  • You can continue self...

How I can do this using fossil (playing guitar)? Export-import are not really option here (also because of the disaster, that I already told)...
And now, think about that I'm not alone and have several colleagues, that have its own origins, but also corporate one, where there have admittedly fewer permissions as I, but the timeline resp. history is nevertheless very long).

Then please, think again about all my another arguments in the previous mail...

 
B.T.W.: I play guitar, but also 4 (or 5) instruments (also hg, and older like svn/cvs)... :)

Regards,
Sergey.

Am 02.06.2017 12:12, schrieb Jan Nijtmans:

2017-06-02 0:37 GMT+02:00 Dipl. Ing. Sergey G. Brester:
But even if we speak about... Another thing, what always hold me back (resp. disturb by the work and cost time) is ... FOSSIL.
Well, your complicated 3-way merge already indicated to me that you're
not familiar with the Fossil way of working. It's different from GIT's, which
- indeed - makes cooperation between the two not trivial (it can be done,
but I wouldn't use import/export for that).

So, you learned to play the piano, but now you are playing the guitar.
Both have strings, both result in beautiful music. It's not easy to play
guitar music on the piano or reverse, unless you are familiar with
both models.

Well, beleave me, playing Fossil is MUCH simpler than playing GIT.
But - indeed - it's different.

Regards,
       Jan Nijtmans

B.T.W.: I play both the guitar and the piano.   ;-)

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Tcl-Core mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/tcl-core
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fossil. [Was: Intent to work on speedups to `clock format` and `clock scan` (#4)]

Larry McVoy-2
His arguments are perfectly valid.  I hate git more than anyone on
this list, based on my knowledge as an SCM creator (NSElite, Teamware,
BitKeeper).  Git sucks, it has no file object, only a tarball object.

That said, even though my baby, BitKeeper is open source, I would not
argue that Tcl should use BitKeeper (even though its three way merge
is pretty sweet).  BitKeeper, like Fossil, is a backwater.  Good luck
finding people who can play either like a guitar or a piano.  But
Git, for all that it sucks, is the the SCM.  It won.  I hate it, I
hate it more than any of you.  But if I were to choose what Tcl
should be under, I'd choose Git.  It's what the world uses.

On Fri, Jun 02, 2017 at 01:23:29PM +0200, Dipl. Ing. Sergey G. Brester wrote:

>  
>
> These were not really the arguments to all things from my statement, or?
>
> My "complicated" 3-way merge was at all not complicate. What makes you
> think that?
> It was just made (even to save my own time), because I had previously
> developed in trunk, and back-ported to 8.6 from trunk (with resolving
> conflicts and changing of all incompatible, e.g. "int refCount" to
> "size_t refCount", etc.), and then as it was ready in 8.6 merged it
> again in trunk (then without conflicts anymore, out-of-the-box).
>
> I known what it in fossil is different from GIT's. Believe me...
> I spoke about another things there... If you don't understand (or rather
> don't want to understand) this, it's good to me. No problem. But don't
> try to explain it all as "You can just not use fossil"...
> Because I'm familiar with both. Just the reasons that were mentioned are
> other as yours...
>
> As an example (mentioned in my previous mail):
>
> I'm on develop-state in the customer-repo. This is proprietary (not
> public). This is an complex project in application server and there tcl,
> tcl-threads. etc. are just a sub-modules (in sense of git).
>
>   * For example, I found some bug (e. g. SEGFAULT), I don't know where
> exactly.
>   * By trying to find it I fixed it directly in modified tcl-code and
> then notice that the core since 8.5 would be affected (but very hard to
> reproduce there).
>   * I'm currently in 8.6-mod-X in foreign origin.
>   * I want to rebase my (extensive) changes to branch based on
> core-8-5-branch.
>   * Then merge "sebres-8-5-cr105" into core-8-6-branch / trunk.
>   * Then reintegrate it into several development stands of some
> customers, to automatically deploy and start the test-cases (I'm lazy
> and have no time at all to do it manually).
>   * Then I found a bug now in my bug-fix.
>   * I'll repeat whole process (7x, so cr105-v.1 up to cr105-v.7).
>   * I have noticed, that the 5th solution was the best one.
>   * Then I'm ready and want to push reintegrated changes back to origin
> (without make it public, because proprietary).
>   * Then I'll reintegrate to origin of tcl.core, I notice the broken
> test-case there. I repair it and repeat this again.
>   * Then I'll reintegrate to origin of tcl.core, to make my changes
> finally public (in the meantime "cr105-v.8", without proprietary code
> pieces, etc.).
>   * You can continue self...
>
> How I can do this using fossil (playing guitar)? Export-import are not
> really option here (also because of the disaster, that I already
> told)...
> And now, think about that I'm not alone and have several colleagues,
> that have its own origins, but also corporate one, where there have
> admittedly fewer permissions as I, but the timeline resp. history is
> nevertheless very long).
>
> Then please, think again about all my another arguments in the previous
> mail...
>
> B.T.W.: I play guitar, but also 4 (or 5) instruments (also hg, and older
> like svn/cvs)... :)
>
> Regards,
> Sergey.
>
> Am 02.06.2017 12:12, schrieb Jan Nijtmans:
>
> > 2017-06-02 0:37 GMT+02:00 Dipl. Ing. Sergey G. Brester:
> >
> >> But even if we speak about... Another thing, what always hold me back (resp. disturb by the work and cost time) is ... FOSSIL.
> >
> > Well, your complicated 3-way merge already indicated to me that you're
> > not familiar with the Fossil way of working. It's different from GIT's, which
> > - indeed - makes cooperation between the two not trivial (it can be done,
> > but I wouldn't use import/export for that).
> >
> > So, you learned to play the piano, but now you are playing the guitar.
> > Both have strings, both result in beautiful music. It's not easy to play
> > guitar music on the piano or reverse, unless you are familiar with
> > both models.
> >
> > Well, beleave me, playing Fossil is MUCH simpler than playing GIT.
> > But - indeed - it's different.
> >
> > Regards,
> > Jan Nijtmans
> >
> > B.T.W.: I play both the guitar and the piano. ;-)
>  

> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot

> _______________________________________________
> Tcl-Core mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/tcl-core


--
---
Larry McVoy                 lm at mcvoy.com             http://www.mcvoy.com/lm 

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Tcl-Core mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/tcl-core
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fossil. [Was: Intent to work on speedups to `clock format` and `clock scan` (#4)]

Brian Griffin-4

On Jun 2, 2017, at 7:55 AM, Larry McVoy <[hidden email]> wrote:

His arguments are perfectly valid.  I hate git more than anyone on
this list, based on my knowledge as an SCM creator (NSElite, Teamware,
BitKeeper).  Git sucks, it has no file object, only a tarball object.

That said, even though my baby, BitKeeper is open source, I would not
argue that Tcl should use BitKeeper (even though its three way merge
is pretty sweet).  BitKeeper, like Fossil, is a backwater.  Good luck
finding people who can play either like a guitar or a piano.  But
Git, for all that it sucks, is the the SCM.  It won.  I hate it, I
hate it more than any of you.  

The MSWindows of the SCM world. :(

But if I were to choose what Tcl
should be under, I'd choose Git.  It's what the world uses.


-Brian


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Tcl-Core mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/tcl-core
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fossil. [Was: Intent to work on speedups to `clock format` and `clock scan` (#4)]

Larry McVoy-2
On Fri, Jun 02, 2017 at 03:02:24PM +0000, Brian Griffin wrote:
> Git, for all that it sucks, is the the SCM.  It won.  I hate it, I
> hate it more than any of you.
>
> The MSWindows of the SCM world. :(

I wish it was that good :(
Windows at least knows what an inode is.  Git does not.

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Tcl-Core mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/tcl-core
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Fossil. [Was: Intent to work on speedups to `clock format` and `clock scan` (#4)]

Peter da Silva-2
At least it’s easier to recover from a catastrophic failure in git.

I don’t think I’ve *ever* reinstalled Windows and got exactly what I had the first time around.
 

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Tcl-Core mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/tcl-core
Loading...