On Fri, Dec 01, 2006 at 07:00:48AM -0800, Marc Shapiro wrote:
> >
> I have never used a VCS before, but I have a question about them. I
> just got a programming job again (after about 13 years). I hate to
> admit, but it is on M$ using VB 6. The company uses Source Safe, which
> has what I consider to be a really big flaw and I am hoping to verify
> that the standard vcs's in linux manage this problem better.
>
I am not sure about Source Safe, having never used it. However, I
personally a number of people who have used it and the *only* positive
thing I have heard about it is that it is well integrated with the MS
toolchain.

> This project contains over 50 classes, a similar number of forms, and
> dozens of non-class modules. Obviously there are a lot of
> interconnections between all of these files. The problem is this: When
> you check out a file the system gives you a fresh, up-to-date copy of
> the file, but tells you nothing about changes to other files that the
> file checked out file is dependant on that my have changed. This means
> there could easily be dozens of other files that have changes in them
> that you actually need to update on your system before you will be able
> to compile and run the program. You find out about these only when you
> do try to run the program and get errors due to changed funtion
> signatures, missing functions and simply changed funtionality within
> existing funtions called from the checked out file but residing in other
> files.
>
> Please tell me that linux's vcs systems handle this better by keeping
> track of file dependancies.
>
What you want is not possible in the current breed of VCS systems. The
best practice is to always update your entire working copy. For it to
know what depends on what would require that it reimplement make. It
would also require that the developer keep that information in some form
that the VCS can understand. It is probably too much of a burden for
most teams.

Regards,

-Roberto

--
Roberto C. Sanchez
http://people.connexer.com/~roberto
http://www.connexer.com

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFFcH2w1snWssAFC08RAlmVAKCYKjprjr/BY5J/CrIsnxElRRSWTACfS75w
JdBsU3msB4OqK0HHBMOAFgM=
=Gw65
-----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 12/01/06 12:30, Nate Duehr wrote:
> Roberto C. Sanchez wrote:
>> On Thu, Nov 30, 2006 at 02:30:54PM -0600, Ron Johnson wrote:
>>> The *real* killer "app" was Linus' decision to develop Linux openly.
>>>
>> I took a software engineering class where the professor maintained that
>> the only notable contribution that Linus Torvalds has made to the
>> programming/compsci/compeng world was figuring out how to make it
>> possible for hundreds of people to work on the same code base without
>> stepping all over each other.
>
> He was wrong. People step all over each other in the kernel and just
> about every application that requires more than a few developers all the
> time in the open-source world. [Hint, see recent ABI screw-ups in
> mysql-server and mysql-client... not caused by Debian, happened
> upstream. Retarded bugs really, too.]

Just because Linus solved a *social* problem, doesn't mean that that
ability was instantly transmitted to every other project.

> CVS/Subversion/[insert tool du jour here] are what REALLY allow
> collaborative development. That and communication, both public and
> private, between the devs.
>
> The fact that "stepping on" happens every day, is why new versions
> aren't released the second the source hits the tree.

- --
Ron Johnson, Jr.
Jefferson LA USA

Is "common sense" really valid?
For example, it is "common sense" to white-power racists that
whites are superior to blacks, and that those with brown skins
are mud people.
However, that "common sense" is obviously wrong.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)

iD8DBQFFcIalS9HxQb37XmcRAkrVAKC2FDXw0aNq22bWOr7/gnawpeseywCg3iBc
O78b6veUMzNwC/ODhhRi6UQ=
=o3bI
-----END PGP SIGNATURE-----


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 12/01/06 12:31, Nate Duehr wrote:
> Francis Healy wrote:
>> The classic definition of the killer app is the one program that
>> justifies the entire cost of the computer.
>
> NICE answer. Wish more people in business would figure that one out.

They did. It was called /Lotus 1-2-3/. A *huge* productivity boost.

- --
Ron Johnson, Jr.
Jefferson LA USA

Is "common sense" really valid?
For example, it is "common sense" to white-power racists that
whites are superior to blacks, and that those with brown skins
are mud people.
However, that "common sense" is obviously wrong.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)

iD8DBQFFcIYOS9HxQb37XmcRAmG0AJ0Qmus1K9c8dzmbk8hDN2 Hi/4rAJACcD55J
JA4pxx8ePVrIjgv6rsd2esE=
=A9it
-----END PGP SIGNATURE-----


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 12/01/06 12:32, Nate Bargmann wrote:
> * Ron Johnson <ron.l.johnson@cox.net> [2006 Dec 01 12:08 -0600]:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> On 12/01/06 10:55, Hugo Vanwoerkom wrote:
>>> hendrik@topoi.pooq.com wrote:
>>>> On Fri, Dec 01, 2006 at 02:54:12PM +0000, George Borisov wrote:
>>>>> Ron Johnson wrote:
>> [snip]
>>> 2. Satisfying your ISP who only install on a M$ OS.
>> ????
>
[snip]
> Now that the temp is around 35, it's working better, but not what it
> should be. But, I don't know how you blow through contracted support
> unless you play the game.

Where do you live? An igloo?

- --
Ron Johnson, Jr.
Jefferson LA USA

Is "common sense" really valid?
For example, it is "common sense" to white-power racists that
whites are superior to blacks, and that those with brown skins
are mud people.
However, that "common sense" is obviously wrong.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)

iD8DBQFFcIbiS9HxQb37XmcRAhqwAKC3Rq1M56ELJMjkcJ/TGt60gt6xMQCfegNn
S5T3RHtnx7JlAjaiBGitJjY=
=higM
-----END PGP SIGNATURE-----


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Marc Shapiro <mshapiro_42@yahoo.com> writes:

> I have never used a VCS before, but I have a question about them. I
> just got a programming job again (after about 13 years). I hate to
> admit, but it is on M$ using VB 6. The company uses Source Safe,
> which has what I consider to be a really big flaw and I am hoping to
> verify that the standard vcs's in linux manage this problem better.
>
> This project contains over 50 classes, a similar number of forms, and
> dozens of non-class modules. Obviously there are a lot of
> interconnections between all of these files. The problem is this:
> When you check out a file the system gives you a fresh, up-to-date
> copy of the file, but tells you nothing about changes to other files
> that the file checked out file is dependant on that my have changed.
> This means there could easily be dozens of other files that have
> changes in them that you actually need to update on your system before
> you will be able to compile and run the program. You find out about
> these only when you do try to run the program and get errors due to
> changed funtion signatures, missing functions and simply changed
> funtionality within existing funtions called from the checked out file
> but residing in other files.
>
> Please tell me that linux's vcs systems handle this better by keeping
> track of file dependancies.

Well, not exactly. CVS only does vcs-per-file. That is, it doesn't
know anything about the interrelationship between files.
The new generation vcs', like subversion and arch does keep track of
batches of dependencies. That is, say you change file A and B (which
depends on each other), and check them in together, then in Subversion,
they have the same version number. So, if you check out by version
number, you will get all the files (and older) that got checked in for
that version.

OTOH, if you are checking out a specific file, I don't know any vcs that
will check out all files that this specific file depend on.

--
John L. Fjellstad
web: http://www.fjellstad.org/ Quis custodiet ipsos custodes


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Fri, Dec 01, 2006 at 11:45:04AM -0800, John L Fjellstad wrote:
>
> Well, not exactly. CVS only does vcs-per-file. That is, it doesn't
> know anything about the interrelationship between files.
> The new generation vcs', like subversion and arch does keep track of
> batches of dependencies. That is, say you change file A and B (which
> depends on each other), and check them in together, then in Subversion,
> they have the same version number. So, if you check out by version
> number, you will get all the files (and older) that got checked in for
> that version.
>
> OTOH, if you are checking out a specific file, I don't know any vcs that
> will check out all files that this specific file depend on.
>

Actually, subversion versions the entire repository.

Regards,

-Roberto

--
Roberto C. Sanchez
http://people.connexer.com/~roberto
http://www.connexer.com


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Ron Johnson wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 12/01/06 12:30, Nate Duehr wrote:
>> Roberto C. Sanchez wrote:
>>> On Thu, Nov 30, 2006 at 02:30:54PM -0600, Ron Johnson wrote:
>>>> The *real* killer "app" was Linus' decision to develop Linux openly.
>>>>
>>> I took a software engineering class where the professor maintained that
>>> the only notable contribution that Linus Torvalds has made to the
>>> programming/compsci/compeng world was figuring out how to make it
>>> possible for hundreds of people to work on the same code base without
>>> stepping all over each other.
>> He was wrong. People step all over each other in the kernel and just
>> about every application that requires more than a few developers all the
>> time in the open-source world. [Hint, see recent ABI screw-ups in
>> mysql-server and mysql-client... not caused by Debian, happened
>> upstream. Retarded bugs really, too.]
>
> Just because Linus solved a *social* problem, doesn't mean that that
> ability was instantly transmitted to every other project.

What social problem did he "solve"? You appear to have him on a
pedestal he probably doesn't deserve nor want.

Back then source was generally free for a whole lot of OS's. He just
wrote a interesting new monolithic kernel for x86 hardware and invited
the world to him work on it. He's no genius of social sciences or
anything.

The rest was just dumb luck and timing. The time was right for
something new, maybe. BSD was going strong by the time Linux popped up.

You give him too much credit. Where he might be a genius is in keeping
it together all these years... not in doing Linux in the first place.

The first Linux kernels weren't exactly earth-shatteringly great or
anything. There were lots of OS's that did a better job on x86
hardware, and a few were already open-source.

Just as an example, Microware's OS/9 is still around, and it was up and
working on multiple hardware platforms years before the early Linux
kernels came out. It never was open-source, but it's still a better
RTOS than Linux.

Nate


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

On Fri, Dec 01, 2006 at 02:50:25PM -0700, Nate Duehr wrote:
>
> The rest was just dumb luck and timing. The time was right for
> something new, maybe. BSD was going strong by the time Linux popped up.
>
According to many who were there, the main reason that Linux got a big
push was because many folks were afraid of the threat by AT&T to the
*BSD variants.

Regards,

-Roberto

--
Roberto C. Sanchez
http://people.connexer.com/~roberto
http://www.connexer.com

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFFcKk75SXWIKfIlGQRAt03AKDCFKVTohZ7aB1HcxetX4 7M+j4J+QCeOF6D
BMDtl6HmKByzkAtPVZRvxV8=
=A1iD
-----END PGP SIGNATURE-----