You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by Neels Janosch Hofmeyr <ne...@elego.de> on 2009/02/25 21:31:48 UTC
Questions from the Subversion 1.6 Preview event in Berlin
Hi all,
Tuesday saw the first day of the Subversion 1.6 Preview (in Berlin),
organized by CollabNet and elego. Everything went well, and there were a lot
of questions about new features and the immediate future plans.
The committers present were Julian Foad, Stephen Butler, Stefan Sperling and
me, and there were 20-odd attendees with corporate background.
So I'd like to pass on some feedback and am hoping for comments that we can
forward back to those guys who asked the questions.
wc-ng
~~~~~
There was a sort of eerie concern about wc-ng, in that it will change some
basic properties of working copies on disk. There was mention of some people
zipping whole working copies and sending them by Email, or storing them away
for later committing. Our general reply to those remarks was "use branches",
but people were insistent that some users will not be happy with making the
use of working copies any more complex by adding restrictions. Needing to
announce the rename or deletion of a working copy to a central database
could, in essence, break a lot of people's workflow, blocking an upgrade.
After stsp broke down the main reasons for and benefits of wc-ng (developer
sanity, speed, concurrency, complex queries), there was some sort of "ok, we
understand now". But, ...
user participation
~~~~~~~~~~~~~~~~~~
... directly from that followed an almost accusingly toned plea to let the
users have a say in the planning of such features like wc-ng. Obviously, it
sounds outright ridiculous to say that to an open source community. But the
reality is that the typical subversion user is not able to keep up with the
dev@ mailing list to filter out any significant changes emerging. Maybe most
of the plans aren't even written down anywhere, but are incubating in the
developers' heads.
So maybe we need some justification for and an estimate of consequences of
(particularly wc-ng, and maybe other) changes that are still in the
pipeline, published on tigris.org, easy to find.
And an affinity to officially released prototypes for user feedback was
mentioned.
other questions
~~~~~~~~~~~~~~~
- We were again asked about svn obliterate. How far are the plans?
Is it going to come any time soon?
- We were again asked about server-imposed rights restrictions.
This time the wish was to "at least" be able to forbid certain
operations in certain directories. Our gut answer was:
use commit hooks. Anything to add here?
- We were asked about automatic commits of svn:externals (directories)
together with their parent directories in the working copy. Can this
be switched on somehow?
- And interest was mentioned in *sparse* svn:externals, to be able
to include a directory to a limited depth as an external.
So, here's to us to answer the outcry of our corporate users. I'll gather
our comments and forward them directly to the participants of the Subversion
1.6 Preview, labeled "comments from the dev@ list".
Thanks
~Neels
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1228969
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Greg Stein <gs...@gmail.com>.
On Thu, Feb 26, 2009 at 06:52, John Mark <jo...@johnmark.org> wrote:
> On Wed, Feb 25, 2009 at 8:24 PM, Justin Erenkrantz
> <je...@apache.org> wrote:
>>
>> I'll repeat my earlier comments - yet again, we're having "community"
>> events being organized with zero notification and collaboration with
>> the actual community. -- justin
>
> I understand where you're coming from. I would also like to have known
> about this event - like, before it took place.
>
> In the past, where would these types of event announcements go? I'd
> like to create a permanent space for this type of activity to go -
> someplace where anyone can announce their event or participate in
> other events. I realize that sending an announcement to a mailing list
> is a good start and is already possible, but surely there are plenty
> of other parties with skin in the game who would be interested in
> planning, sponsoring and attending events. In that case, we
> (CollabNet) are happy to facilitate that.
As a start, just send something to the dev@ and users@ mailing lists.
Don't try to make something big and difficult, or you'll never get it
off the ground. Start with a simple email. That's way better than what
we appear to have today.
Cheers,
-g
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1231207
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by John Mark <jo...@johnmark.org>.
On Wed, Feb 25, 2009 at 8:24 PM, Justin Erenkrantz
<je...@apache.org> wrote:
>
> I'll repeat my earlier comments - yet again, we're having "community"
> events being organized with zero notification and collaboration with
> the actual community. -- justin
I understand where you're coming from. I would also like to have known
about this event - like, before it took place.
In the past, where would these types of event announcements go? I'd
like to create a permanent space for this type of activity to go -
someplace where anyone can announce their event or participate in
other events. I realize that sending an announcement to a mailing list
is a good start and is already possible, but surely there are plenty
of other parties with skin in the game who would be interested in
planning, sponsoring and attending events. In that case, we
(CollabNet) are happy to facilitate that.
John Mark Walker
Community Manager, openCollabNet
http://www.collab.net/
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1231142
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by "C. Michael Pilato" <cm...@collab.net>.
Justin Erenkrantz wrote:
> Yup yup. As I've said before, such events would only be better served
> by the real community knowing about it. -- justin
Ah, okay. Thanks for the clarification of the Bi-Justinian Counsel. :-)
I'm still a little put off by references to the "real community", as if only
those folks who have the time and ability to follow our highly active
mailing lists matter, but ... *shrug* I'm already over it.
I agree that to the extent possible and practical, the success of
Subversion-related events (if success is determined by the participation of
the folks most likely to know the code and its trajectory intimately) can
only be helped by letting more of us list-folk know about these events
before they occur. And even if we aren't invited to, or cannot, attend, we
(the devs) all want to ensure that when folks speak for us -- as they
ultimately do when they talk about soon-to-be-released features and current
development work -- they do so accurately.
--
C. Michael Pilato <cm...@collab.net>
CollabNet <> www.collab.net <> Distributed Development On Demand
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1233087
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Justin Erenkrantz <je...@apache.org>.
On Wed, Feb 25, 2009 at 10:06 PM, Greg Stein <gs...@gmail.com> wrote:
> I'm assuming Justin is like myself in thinking, "shit. that woulda
> been neat to know about and possibly attend." ... NOT "damn. that
> shouldn't have happened without our permission." Your entire reply is
> as if Justin said it *should not* have happened, when I suspect is was
> more like *why couldn't the rest of us have been involved?*
As usual, Greg channels me quite well. =)
> I mean... this even has a Capitalized Name. And apparently Tuesday was
> just the first day.
>
> Or maybe it was just for elego and CN customers. Fine. Cool. But when
> the first line of an email talks about an Event, then my first thought
> is "huh? what event? why wasn't I invited? I like SVN. I think I could
> have contributed in some way. I woulda been interested to talk to
> these people." (and after reading the report about wc-ng, I think it
> was mischaracterized, so I *really* would have wished to have been
> there)
Exactly my feelings.
> So all that said, your email is right in that we cannot and should not
> attempt to control any happenings around Subversion. But I suspect
> you're preaching to the choir, and Justin wasn't saying that at all.
> It just seems weird when there is an event that seems/appears/sounds
> like a "community" event, and we have NO IDEA that it is happening. I
> cannot imagine any community event that would NOT want our
> participation (our == the devs), so why aren't we hearing about these
> things?
Yup yup. As I've said before, such events would only be better served
by the real community knowing about it. -- justin
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1231417
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Greg Stein <gs...@gmail.com>.
On Fri, Feb 27, 2009 at 23:41, Neels J Hofmeyr <ne...@elego.de> wrote:
>...
> I personally don't oppose separate groups "conspiring" about Subversion,
> giving it a fancy name and sending a resume of their insights to the dev
> list. You might argue about that. Regardless, the "Subversion 1.6 Preview"
> was surely *not* intended to be such a "conspiration". I'm sorry to hear
> that it turned out to make that impression.
nono.... I don't think *anybody* was worried about "conspiring'. Certainly not.
People are just confused that events happen that they don't even hear about it.
Cheers,
-g
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1243606
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Neels Janosch Hofmeyr <ne...@elego.de>.
Hi again,
first of all thanks for the feedback!
I'm sorry to hear that virtually no-one knew about this event. I had assumed
the organizers would have thought of the single most obvious place to
announce any event around Subversion development. dev@subversion.tigris.org.
That surely was a pretty stupid mistake. Duh. I'll remind our guys of it.
It was two one-day events, pretty much the same thing staged in Berlin on
Tuesday and repeated in Munich on Wednesday. See the agendas here:
http://www.elegosoft.com/en/company/start/svnpreview.html
(I was asked to attend in Berlin and present some of the slides. I ended up
answering questions to the best of my ability and noted unsatisfactorily
answered ones down on paper, to ask the list about them ... assuming that
everyone was in the know about the event in the first place!)
I personally don't oppose separate groups "conspiring" about Subversion,
giving it a fancy name and sending a resume of their insights to the dev
list. You might argue about that. Regardless, the "Subversion 1.6 Preview"
was surely *not* intended to be such a "conspiration". I'm sorry to hear
that it turned out to make that impression.
Won't happen again.
~Neels
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1241255
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Stefan Sperling <st...@elego.de>.
On Thu, Feb 26, 2009 at 07:06:02AM +0100, Greg Stein wrote:
> Or maybe it was just for elego and CN customers. Fine. Cool. But when
> the first line of an email talks about an Event, then my first thought
> is "huh? what event? why wasn't I invited? I like SVN. I think I could
> have contributed in some way. I woulda been interested to talk to
> these people."
This event was publicly announced on a few channels, including
the elego website (but no, I don't expect you guys to look at that
site).
I will try to make sure that such events get announced here, too,
in the future (if they happen again).
> (and after reading the report about wc-ng, I think it
> was mischaracterized, so I *really* would have wished to have been
> there)
Don't worry, it was not mischaracterized.
I can send you slide sets if you want to look at them
to get an idea about what speakers said.
I'd say if there is general interest we might even be
able to make slides publicly available (but that is not
for me to decide, so I cannot make promises without consulting
elego/collab.net overlords first).
Note that what Neels wrote was not what we told people, it was
what people asked us about and said to us, having misunderstood
a few points, or simply making assumptions out of the blue,
not based on actual facts. The one user in question who voiced his
concerns about wc-ng being destabilising apparently had the impression
that we were ripping out a known-working subsystem and replacing it
entirely by new code. He admitted that he didn't even know that
the design docs etc. are publicly available. My impression was that
he did not understand how development of wc-ng actually happens.
This was a complete non-issue in my opinion, I don't understand
why this was brought up here on this list.
Also, what Neels presented was only a really small subset
of questions that were asked. He did not mention the many
questions that were asked that all of us hear regularly
on our lists and know the answers to. He also could not
include questions from the second day because he was not there.
Stefan
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Greg Stein <gs...@gmail.com>.
On Thu, Feb 26, 2009 at 06:21, C. Michael Pilato <cm...@collab.net> wrote:
> Justin Erenkrantz wrote:
>> On Wed, Feb 25, 2009 at 1:31 PM, Neels Janosch Hofmeyr <ne...@elego.de> wrote:
>>> Tuesday saw the first day of the Subversion 1.6 Preview (in Berlin),
>>> organized by CollabNet and elego. Everything went well, and there were a lot
>>
>> I'll repeat my earlier comments - yet again, we're having "community"
>> events being organized with zero notification and collaboration with
>> the actual community. -- justin
>
> Justin, I think you're reading something more than is really present.
Now you're reading more than is present :-)
I'm assuming Justin is like myself in thinking, "shit. that woulda
been neat to know about and possibly attend." ... NOT "damn. that
shouldn't have happened without our permission." Your entire reply is
as if Justin said it *should not* have happened, when I suspect is was
more like *why couldn't the rest of us have been involved?*
I mean... this even has a Capitalized Name. And apparently Tuesday was
just the first day.
Or maybe it was just for elego and CN customers. Fine. Cool. But when
the first line of an email talks about an Event, then my first thought
is "huh? what event? why wasn't I invited? I like SVN. I think I could
have contributed in some way. I woulda been interested to talk to
these people." (and after reading the report about wc-ng, I think it
was mischaracterized, so I *really* would have wished to have been
there)
So all that said, your email is right in that we cannot and should not
attempt to control any happenings around Subversion. But I suspect
you're preaching to the choir, and Justin wasn't saying that at all.
It just seems weird when there is an event that seems/appears/sounds
like a "community" event, and we have NO IDEA that it is happening. I
cannot imagine any community event that would NOT want our
participation (our == the devs), so why aren't we hearing about these
things?
Cheers,
-g
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1231199
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by "C. Michael Pilato" <cm...@collab.net>.
Justin Erenkrantz wrote:
> On Wed, Feb 25, 2009 at 1:31 PM, Neels Janosch Hofmeyr <ne...@elego.de> wrote:
>> Tuesday saw the first day of the Subversion 1.6 Preview (in Berlin),
>> organized by CollabNet and elego. Everything went well, and there were a lot
>
> I'll repeat my earlier comments - yet again, we're having "community"
> events being organized with zero notification and collaboration with
> the actual community. -- justin
Justin, I think you're reading something more than is really present.
"The Subversion community" is not the same as "the Subversion development
community". Nor does a "community event" imply or require the consultation
with or participation of any particular subset of either "the Subversion
user community" or "the Subversion developer community". Any place where
two or more people can find some common ground there is "actual community".
If there are some Subversion users in Berlin who've found common interest
in their Subversion usage, that's great. If a person or company can
interact with that community of users for the benefit of the users, for the
benefit of Subversion, and perhaps even for the benefit of that person or
company -- fantastic. Why should the minority of Subversion users who are
developers -- or even the minority of users who are subscribed to users@ --
work to suppress the mutual exchange of information about Subversion,
*especially* when that information can be fed back into this list and used
to further the project?
I, too, was completely unaware that CollabNet and elego had organized some
kind of a Subversion 1.6 preview. Total surprise to me. But what if the
ASF had hosted it? What if Google had done it? What about any one of us
acting individually? What if Blair Zajac, wearing a consultant hat, had
decided to organize a little meeting of folks in his hometown who shared a
common interest in Subversion? What if it wasn't Blair, but a random user
we've never met who's just really fired up about our version control tool?
The Subversion cat is out of the bag; we can't dictate the happenings of the
humongous ecosystem that has formed around it. But to me, that's something
to celebrate, not something to bemoan!
Are you concerned that elego or CollabNet or somebody else is
misrepresenting themselves or Subversion, or doing something on the sly? Is
there some particular concern with the lack of communication with whatever
you've termed "the actual community" besides just that lack of
communication? If so, I'd encourage you to voice those concerns so they can
be better understood and addressed.
--
C. Michael Pilato <cm...@collab.net>
CollabNet <> www.collab.net <> Distributed Development On Demand
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1230997
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Justin Erenkrantz <je...@apache.org>.
On Wed, Feb 25, 2009 at 1:31 PM, Neels Janosch Hofmeyr <ne...@elego.de> wrote:
> Tuesday saw the first day of the Subversion 1.6 Preview (in Berlin),
> organized by CollabNet and elego. Everything went well, and there were a lot
I'll repeat my earlier comments - yet again, we're having "community"
events being organized with zero notification and collaboration with
the actual community. -- justin
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1230831
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by "Hyrum K. Wright" <hy...@mail.utexas.edu>.
On Feb 25, 2009, at 4:10 PM, Karl Fogel wrote:
> Bert Huijben <rh...@sharpsvn.net> writes:
>> The only major difference this introduce is that you can no longer
>> copy a
>> single directory to another location an thread that as a separate
>> working
>> copy. (But this is a far less common operation than moving or just
>> copying
>> an entire working copy).
>
> ...and there will be an 'svn detach' command or something to allow
> that
> too, right?
That's The Plan (though there hasn't been an incredible amount of
discussion about it, yet).
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1229256
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Karl Fogel <kf...@red-bean.com>.
Bert Huijben <rh...@sharpsvn.net> writes:
> The only major difference this introduce is that you can no longer copy a
> single directory to another location an thread that as a separate working
> copy. (But this is a far less common operation than moving or just copying
> an entire working copy).
...and there will be an 'svn detach' command or something to allow that
too, right?
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1229208
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Mark Phippard <ma...@gmail.com>.
On Sun, Mar 1, 2009 at 6:11 PM, Branko Cibej <br...@xbc.nu> wrote:
> Bert Huijben wrote:
>> Note that the default behavior of WC-NG will be to store all administrative
>> area in a single .svn directory in the working copy's root. This will still
>> allow zipping up a repository, etc. just like WC-1.0.
>>
>
> Are you sure you can just zip up some database created by a semi-random
> variant of sqlite3, unpack it and use it on another box with a different
> OS, word size and, gods forbid, endianness?
According to the "Definitive Guide to SQLite" the answer is absolutely
yes. The file is guaranteed to be binary compatible across all
platforms, endianness, 32/64-bit etc..
http://books.google.com/books?id=VsZ5bUh0XAkC&pg=PA8&lpg=PA8&dq=sqlite+portability+of+database+file&source=bl&ots=u45Kmp52F5&sig=2A_Bst88utrWE-QcCNv6JipBrnw&hl=en&ei=HRmrSc_CA-CbtwfD6ODvDw&sa=X&oi=book_result&resnum=3&ct=result
That said, it does seem like different versions of SQLite can
theoretically cause problems:
http://www.sqlite.org/formatchng.html
--
Thanks
Mark Phippard
http://markphip.blogspot.com/
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1253011
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Greg Hudson <gh...@mit.edu>.
(Actually replying to Branko, but I missed his message.)
On Thu, 2009-03-05 at 22:44 +0100, Steinar Bang wrote:
> >>>>> Branko Cibej <br...@xbc.nu>:
>
> > Are you sure you can just zip up some database created by a
> > semi-random variant of sqlite3, unpack it and use it on another box
> > with a different OS, word size and, gods forbid, endianness?
http://www.sqlite.org/fileformat.html#section_ suggests that the file
format is independent of word size and byte order. I don't know what
their policy is across versions. Also, that document doesn't describe
the journal or statement files, so I don't know what would happen if
there's data lingering there at the time of transfer.
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1277911
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Mark Phippard <ma...@gmail.com>.
On Thu, Mar 5, 2009 at 4:44 PM, Steinar Bang <sb...@dod.no> wrote:
>>>>>> Branko Cibej <br...@xbc.nu>:
>
>> Are you sure you can just zip up some database created by a
>> semi-random variant of sqlite3, unpack it and use it on another box
>> with a different OS, word size and, gods forbid, endianness?
>
> Isn't the sqlite3 stuff just a cache? Or is everything in wc-ng inside
> an sqlite3 db?
It contains the metadata and properties. Basically the stuff that is
in entries and the properties. It does not contain the text-base.
--
Thanks
Mark Phippard
http://markphip.blogspot.com/
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1277836
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Steinar Bang <sb...@dod.no>.
>>>>> Branko Cibej <br...@xbc.nu>:
> Are you sure you can just zip up some database created by a
> semi-random variant of sqlite3, unpack it and use it on another box
> with a different OS, word size and, gods forbid, endianness?
Isn't the sqlite3 stuff just a cache? Or is everything in wc-ng inside
an sqlite3 db?
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1273902
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Branko Cibej <br...@xbc.nu>.
Bert Huijben wrote:
> Note that the default behavior of WC-NG will be to store all administrative
> area in a single .svn directory in the working copy's root. This will still
> allow zipping up a repository, etc. just like WC-1.0.
>
Are you sure you can just zip up some database created by a semi-random
variant of sqlite3, unpack it and use it on another box with a different
OS, word size and, gods forbid, endianness?
-- Brane
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1252954
RE: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Bert Huijben <rh...@sharpsvn.net>.
> -----Original Message-----
> From: Neels Janosch Hofmeyr [mailto:neels@elego.de]
> Sent: Wednesday, February 25, 2009 10:32 PM
> To: dev@subversion.tigris.org
> Subject: Questions from the Subversion 1.6 Preview event in Berlin
>
> Hi all,
>
> Tuesday saw the first day of the Subversion 1.6 Preview (in Berlin),
> organized by CollabNet and elego. Everything went well, and there were
> a lot
> of questions about new features and the immediate future plans.
>
> The committers present were Julian Foad, Stephen Butler, Stefan
> Sperling and
> me, and there were 20-odd attendees with corporate background.
>
> So I'd like to pass on some feedback and am hoping for comments that we
> can
> forward back to those guys who asked the questions.
>
>
> wc-ng
> ~~~~~
>
> There was a sort of eerie concern about wc-ng, in that it will change
> some
> basic properties of working copies on disk. There was mention of some
> people
> zipping whole working copies and sending them by Email, or storing them
> away
> for later committing. Our general reply to those remarks was "use
> branches",
> but people were insistent that some users will not be happy with making
> the
> use of working copies any more complex by adding restrictions. Needing
> to
> announce the rename or deletion of a working copy to a central database
> could, in essence, break a lot of people's workflow, blocking an
> upgrade.
>
> After stsp broke down the main reasons for and benefits of wc-ng
> (developer
> sanity, speed, concurrency, complex queries), there was some sort of
> "ok, we
> understand now". But, ...
Note that the default behavior of WC-NG will be to store all administrative
area in a single .svn directory in the working copy's root. This will still
allow zipping up a repository, etc. just like WC-1.0.
(There are plans for allowing the user to move the data entirely out of the
working copy but this won't be the default behavior, as there is no always
valid default location where we can move everything to).
The only major difference this introduce is that you can no longer copy a
single directory to another location an thread that as a separate working
copy. (But this is a far less common operation than moving or just copying
an entire working copy).
Bert
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1229068
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Neels Janosch Hofmeyr <ne...@elego.de>.
Neels Janosch Hofmeyr wrote:
> Hi all,
>
> Tuesday saw the first day of the Subversion 1.6 Preview (in Berlin),
> organized by CollabNet and elego. Everything went well, and there were a lot
> of questions about new features and the immediate future plans.
>
> The committers present were Julian Foad, Stephen Butler, Stefan Sperling and
> me, and there were 20-odd attendees with corporate background.
Correction, there were 11 in Berlin.
~Neels
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1241816
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Neels Janosch Hofmeyr <ne...@elego.de>.
Neels Janosch Hofmeyr wrote:
> - And interest was mentioned in *sparse* svn:externals, to be able
> to include a directory to a limited depth as an external.
There's an issue for that: #3216
~Neels
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1229023
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by John Mark <jo...@johnmark.org>.
On Wed, Feb 25, 2009 at 2:41 PM, Karl Fogel <kf...@red-bean.com> wrote:
> Neels had a good point. For complex but important topics like the wc
> rewrite, a summary now and then would make a big difference to the many
> users who don't have time to follow the development list.
>
> Lots of projects keep a blog for this sort of thing. Subversion never
> has, but it might be time to consider one. Or if the Submerged blog at
> http://blogs.open.collab.net/svn/ is doing that job already, then link
> to it from somewhere on the front page.
I hope nobody minds if I chime in here... but I think this is a great
point and, in my opinion, one thing that's missing from Subversion. As
I put together more community days - more on that later - I'd love to
know where to send the people we meet so they can find topical info on
new developments. I'm happy to help formulate a system that we can use
for just such a purpose. One thing I'm about to implement is importing
feeds into the submerged blog. That way, whenever one of you posts
something interesting, I'll add it there.
The submerged blog *could* be something that fills this need, but
we've let it lie still for too long - not pointing fingers, because
I'm just as guilty as anybody else. I'd like to remedy that ASAP.
Considering the amount of "mindshare" grabbed by other SCM's within
various open source dev communities, I think now is past time to look
at renewed outreach efforts, with more blogging being just one.
If you'd like to brainstorm things we could do in concert, I'm happy
to drop in on IRC and have that discussion.
Thanks,
JM
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1229428
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Karl Fogel <kf...@red-bean.com>.
Justin Erenkrantz <ju...@erenkrantz.com> writes:
> On Wed, Feb 25, 2009 at 2:41 PM, Karl Fogel <kf...@red-bean.com> wrote:
>> (By the way, we've not had a really good plan for what to do with
>> donations to the SVN Corp. We'd probably get a lot more donations if we
>> did; it might work for SVN Corp to fund development, just would have to
>> be thought through carefully...)
>
> Besides reiterating my belief that SVN Corp shouldn't even exist at
> all as a separate entity, I would be opposed to having SVN Corp
> funding development. My experience with open-source organizations
> trying to fund development has been a big fat zero with lots of
> negative follow-on effects for the community. -- justin
Okay.
Note that this doesn't stop anyone else who wants to organize
collaborative funding for some development effort -- and I'd certainly
encourage anyone to do that (and donors to donate to it).
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1230918
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Justin Erenkrantz <je...@apache.org>.
On Wed, Feb 25, 2009 at 2:41 PM, Karl Fogel <kf...@red-bean.com> wrote:
> (By the way, we've not had a really good plan for what to do with
> donations to the SVN Corp. We'd probably get a lot more donations if we
> did; it might work for SVN Corp to fund development, just would have to
> be thought through carefully...)
Besides reiterating my belief that SVN Corp shouldn't even exist at
all as a separate entity, I would be opposed to having SVN Corp
funding development. My experience with open-source organizations
trying to fund development has been a big fat zero with lots of
negative follow-on effects for the community. -- justin
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1230859
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Karl Fogel <kf...@red-bean.com>.
"Hyrum K. Wright" <hy...@mail.utexas.edu> writes:
> Along the lines of not paying attention to users, did this arise from
> the "Speeding up workspace" thread on the users list, perchance? This
> is open source, people: if you want to have influence in the
> community, contribute. Period. (Caveat: "contribute" does not mean
> "write code". It could mean write code, but it could also mean join
> in the development discussion, fund a developer to work on your
> feature, or contribute financially to SVNCorp. But being visible in
> the community will give you a much louder voice when it comes to
> features, etc.)
I know I'm totally kibbitzing, having not been active in development
lately, but:
Neels had a good point. For complex but important topics like the wc
rewrite, a summary now and then would make a big difference to the many
users who don't have time to follow the development list.
Lots of projects keep a blog for this sort of thing. Subversion never
has, but it might be time to consider one. Or if the Submerged blog at
http://blogs.open.collab.net/svn/ is doing that job already, then link
to it from somewhere on the front page.
Most users already know that increased investment of time & effort will
bring increased influence. No one thinks that's unjust or anything.
But it's still good (for the project as well as for users) to provide
important stuff in a way that's easy for non-involved users to follow.
(By the way, we've not had a really good plan for what to do with
donations to the SVN Corp. We'd probably get a lot more donations if we
did; it might work for SVN Corp to fund development, just would have to
be thought through carefully...)
-Karl
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1229375
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by Neels Janosch Hofmeyr <ne...@elego.de>.
Hyrum K. Wright wrote:
>
> On Feb 25, 2009, at 3:31 PM, Neels Janosch Hofmeyr wrote:
>>
>> - We were asked about automatic commits of svn:externals (directories)
>> together with their parent directories in the working copy. Can this
>> be switched on somehow?
>
> Are people concerned that this is happening? Or is this a feature request?
It's a feature request.
I guess.
It didn't sound that urgent, it's just something I'd like to follow up on.
So is there currently no way to have dir externals automatically commit with
their parent folders?
~Neels
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1241686
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by "Hyrum K. Wright" <hy...@mail.utexas.edu>.
On Feb 25, 2009, at 3:31 PM, Neels Janosch Hofmeyr wrote:
> Hi all,
>
> Tuesday saw the first day of the Subversion 1.6 Preview (in Berlin),
> organized by CollabNet and elego. Everything went well, and there
> were a lot
> of questions about new features and the immediate future plans.
>
> The committers present were Julian Foad, Stephen Butler, Stefan
> Sperling and
> me, and there were 20-odd attendees with corporate background.
>
> So I'd like to pass on some feedback and am hoping for comments that
> we can
> forward back to those guys who asked the questions.
>
>
> wc-ng
> ~~~~~
>
> There was a sort of eerie concern about wc-ng, in that it will
> change some
> basic properties of working copies on disk. There was mention of
> some people
> zipping whole working copies and sending them by Email, or storing
> them away
> for later committing. Our general reply to those remarks was "use
> branches",
> but people were insistent that some users will not be happy with
> making the
> use of working copies any more complex by adding restrictions.
> Needing to
> announce the rename or deletion of a working copy to a central
> database
> could, in essence, break a lot of people's workflow, blocking an
> upgrade.
>
> After stsp broke down the main reasons for and benefits of wc-ng
> (developer
> sanity, speed, concurrency, complex queries), there was some sort of
> "ok, we
> understand now". But, ...
Users with WC-NG won't need to announce anything to a central
database. As Bert already mentioned, the database will be central to
a particular working copy as a default, and the working copy will be
able to be shipped around.
>
>
>
> user participation
> ~~~~~~~~~~~~~~~~~~
>
> ... directly from that followed an almost accusingly toned plea to
> let the
> users have a say in the planning of such features like wc-ng.
> Obviously, it
> sounds outright ridiculous to say that to an open source community.
> But the
> reality is that the typical subversion user is not able to keep up
> with the
> dev@ mailing list to filter out any significant changes emerging.
> Maybe most
> of the plans aren't even written down anywhere, but are incubating
> in the
> developers' heads.
We've got quite a corpus of notes on wc-ng, including a development
plan and a meta-issue in the issue tracker. You can point people to http://svn.collab.net/repos/svn/trunk/notes/wc-ng-design
Most other proposed features are likewise doc'd ... somewhere.
Along the lines of not paying attention to users, did this arise from
the "Speeding up workspace" thread on the users list, perchance? This
is open source, people: if you want to have influence in the
community, contribute. Period. (Caveat: "contribute" does not mean
"write code". It could mean write code, but it could also mean join
in the development discussion, fund a developer to work on your
feature, or contribute financially to SVNCorp. But being visible in
the community will give you a much louder voice when it comes to
features, etc.)
> So maybe we need some justification for and an estimate of
> consequences of
> (particularly wc-ng, and maybe other) changes that are still in the
> pipeline, published on tigris.org, easy to find.
>
> And an affinity to officially released prototypes for user feedback
> was
> mentioned.
I used to have a nightly script running on orac which would create
proto-tarballs; I'd be happy to resurrect it for trunk. The only
caveat is that people need to know that it's completely untested code,
with all the caveats that it implies. As a developer, I'm happy to
have people running this code, so long as they understand it might eat
their children.
>
> other questions
> ~~~~~~~~~~~~~~~
>
> - We were again asked about svn obliterate. How far are the plans?
> Is it going to come any time soon?
There's been some rumbling on the list about designs and such, but
zero implementation effort thus far.
> - We were again asked about server-imposed rights restrictions.
> This time the wish was to "at least" be able to forbid certain
> operations in certain directories. Our gut answer was:
> use commit hooks. Anything to add here?
>
> - We were asked about automatic commits of svn:externals (directories)
> together with their parent directories in the working copy. Can this
> be switched on somehow?
Are people concerned that this is happening? Or is this a feature
request?
> - And interest was mentioned in *sparse* svn:externals, to be able
> to include a directory to a limited depth as an external.
>
>
> So, here's to us to answer the outcry of our corporate users. I'll
> gather
> our comments and forward them directly to the participants of the
> Subversion
> 1.6 Preview, labeled "comments from the dev@ list".
Thanks for the writeup. I generally appreciate this type of feedback
from the user community, just so long as they aren't complaining *too*
loudly. :)
-Hyrum
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1229251
Re: Questions from the Subversion 1.6 Preview event in Berlin
Posted by km...@rockwellcollins.com.
> wc-ng
> ~~~~~
>
> There was a sort of eerie concern about wc-ng, in that it will change
some
> basic properties of working copies on disk. There was mention of some
people
> zipping whole working copies and sending them by Email, or storing them
away
> for later committing. Our general reply to those remarks was "use
branches",
> but people were insistent that some users will not be happy with making
the
> use of working copies any more complex by adding restrictions. Needing
to
> announce the rename or deletion of a working copy to a central database
> could, in essence, break a lot of people's workflow, blocking an
upgrade.
>
> After stsp broke down the main reasons for and benefits of wc-ng
(developer
> sanity, speed, concurrency, complex queries), there was some sort of
"ok, we
> understand now". But, ...
As a corporate user I welcome a working copy rewrite. I'm assuming
there is a "central database" for each working copy, so one could still
zip up the directory tree and mail it. I don't think this is as
bad as you made it sound.
> - We were again asked about svn obliterate. How far are the plans?
> Is it going to come any time soon?
This is almost a must for corporate users, since completely dumping
my largest 150G repo, filtering it, and reloading it would probably
be impractical if not impossible.
> - We were again asked about server-imposed rights restrictions.
> This time the wish was to "at least" be able to forbid certain
> operations in certain directories. Our gut answer was:
> use commit hooks. Anything to add here?
The biggest problem is the disconnect between the different
authentication and authorization steps. For example, if you use
apache and ldap, you can easily map group members to repository access.
But if you also need directory level restrictions you then must pull
data out of ldap into a special authz file. If you then need
to restrict certain operations via a hook script, you must then
pull data out of ldap into a special restrictions file. Thus you
now have 3 potentially separate configuration areas that you
must manually keep in sync.
Kevin R.
------------------------------------------------------
http://subversion.tigris.org/ds/viewMessage.do?dsForumId=462&dsMessageId=1229130