[Owncloud] Upgraded to Owncloud 5. Here come the conflict files again

[Simon Brereton]

On 21 March 2013 11:16, hemathor <hemathor at gmx.de> wrote:
> hey,
>
>> Yeah, this is really frustrating.
>
> you don't say!.. i'm totally with you!
>
>> Given that file synchronization is
>> kind of the core feature of this application, [...]
>
> is it? is it really? or do you just think it is, because it is so important to you? (and to me, btw., hence this reply...)
>
> given the amount of mails here in this list, issues with conflict files and with syncing in general are almost nonexistent. or no one seems to care, which i think is the case...
>
>> I really don't
>> understand why these issues have persisted for so long.
>
> i have an idea: to few people are working on that part of owncloud. and the ones who do suffer probably from the same attitude all other free (as in free beer) opensource software developers are dealing with:
>
>   it is so much cooler and exciting to add new functions
>   than making existing ones rocksolid!
>
> no one likes to bugfix ageold software when noone can see the results right away (having new fancy icons, new fancy apps, new fancy whatever, that catches the user's attention).
>
> the developers of the syncinc part of owncloud suffer from that to an enormous extend. i know what i'm talking about, because i myself worked for decades developing software no one sees. some call it middleware, some framework. no one cares or even appreciates, if everything works smoothly and flawless, but as soon as there is the slightest glitch in the matrix, you're the user's most favourite punching ball.
>
>> As is, file
>> synchronization is barely even usable because of this.
>
> agreed! that's why i still stick to dropbox (with encfs) for the important stuff, and only put the unimportant files to owncloud.

That is so sad.

>> I'd love to help make this better in some way.
>
> that's the spirit! :)
>
>> I think the overall focus of
>> OwnCloud is great, and I love that there's an open source application
>> like this.  I just hope it continues to improve.
>
> like with all free (as in free beer) opensource software:
>   if you don't like it: do it yourself!
>   but never ever expect the developers to do what you want...
>
> to say it again and clearly:
> i really do appreciate the work of the developers of owncloud and especially of the syncing team! really!
> but focussing on stability instead of adding more and more functionality would't hurt the slightest bit...

For me, this is the most frustrating thing about the Owncloud project.
 I've been a user since 3.something and I would love to offer my
clients an alternative to other cloud storage systems, but remain
reluctant to do so.  To be clear, the product works for me and my
private use, but I couldn't imagine rolling it out to a few domains
and supporting more than 2-5 *technical* users.  Someone earlier in
the thread mentioned focus, and I bit my tongue, because I am not a
developer and therefore not in a position to add back to the
code/project in anyway.  I had hoped the recent appointment of a PM
would solve some of these issues and it may yet do so, but I haven't
seen it so far.  Traditionally, FOSS projects was about doing one
thing and doing it well.  Owncloud on the other hand, and please take
this as constructive criticism because I truly believe this project
will be huge in the years to come, has no focus and seems to want to
do everything - it won't be long before someone wants to add in a mail
client with some sort of imap-sync because for example you already
have OC syncing your addressbook, calendar and tasks, (and now someone
is asking for bookmarks).  Horde or SoGo already offer all this
functionality, including the mobile aspect, so it's frustrating to see
the wheel being reinvented and reinvented badly, because of the rush
for the next release.

Every release promises more and more features because they are cool
and there's a member of the team prepared to contribute this viewer,
or that app, or whatever.  Very rarely do I see functionality from
every day users and more importantly from admins - who are able to
filter and condense user requested functionality - implemented or
planned.  Instead of trying to build it yourselves, why not reuse
code/modules/functionality from other FOSS projects (so long as the
licence is compatible - but isn't that the point of the GPL?  That you
can just take a whole block of code/functionality and add it to your
project so long as you a) credit it and b) contribute back any
enhancements?

Server-side encryption was surely a must, so I'm happy that has been
planned/implemented, but I would dearly love to see OC mature in it's
roadmap approach, contribute and collaborate with other projects and
spend more time on bugs and "real-world" functionality requests.  I
would love to contribute more bugs and testing - but as per the
comment above, I simply don't use OC enough and so I don't see the
impacts of the new releases that others are seeing (the sync/conflict
issues for example).

There are clever people in this project.  And I learn a lot from
watching Jan and Klaus and Tom construct argument, explain rationale,
and as I said, OC will be come vastly more important as people's
privacy is ever more eroded and attacks on commercial services (i.e.
Dropbox) increase because of their ROI.  I just hope that it's mature
enough to cope with that influx.

Simon